.

Angewandte R. Schlçgl
Reviews

DOI: 10.1002/anie.201410738
Heterogeneous Catalysis Special Issue 150 Years of BASF

Heterogeneous Catalysis
Robert Schlçgl*

Keywords:
heterogeneous catalysis ·
reaction kinetics ·
surface chemistry

Angewandte
Chemie
Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3465
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

A heterogeneous catalyst is a functional material that continually

creates active sites with its reactants under reaction conditions. These
sites change the rates of chemical reactions of the reactants localized
on them without changing the thermodynamic equilibrium between
the materials.

“And yet it is only by studying function that we can understand Table 1: Thermodynamic data for select catalytic oxidation reactions.

function, so that the kinetic aspect must be allowed to retain its Sratting Target Mol water per DH298 DH298
material product Mol starting target prod- Total oxida-
place and assigned its part in the unravelling of the great
material uct tion
mystery ” CH3OH CH2O 1 311 675
C. N. Hinshelwood, 1947
C2H4 C2H4O 0 438 1323
“Measure that which is measurable and make measurable that
C3H6
i-C4H8
C3H4O
C4H6
1
1
365
242
1959
2522
which is not ”
Galileo Galilei

enthalpy for the desired reactions and for the undesired, but
1. Introduction thermodynamically preferred, total oxidation. In all cases, the
reaction product is also less stable than the starting material.
Catalysis is the science and technology of influencing the The Table combines oxidative dehydrogenation, which nec-
rates of chemical reactions. A catalyst is a material that essarily produces water, with the oxidation reactions that
changes the path of a chemical reaction without itself being result with and without compulsory coproducts. The sum of
expended. In this way a small amount of catalyst material can the heats of reaction for total oxidation develop an enormous
convert a large quantity of reactants and this happens driving force with increasing molecule size for the overall
preferentially under milder conditions than would be kinetics and exemplify the directing role of the stability of the
required by the stoichiometric reaction pathway. If more compulsory coproducts.
than one reaction product is possible, the catalyst may change The Table illustrates the large challenges in carrying out
the distribution of these products compared to stoichiometric these processes in a technologically effective way. This applies
conversion and thereby allow control of the selectivity of of course to the selectivity of the catalyst and also to the
a chemical reaction. reaction engineering, which must be able to safely transport
In rare cases, the result of a chemical reaction is only one the heat of reaction away while simultaneously allowing its
product. Here, an acceleration of the reaction rate and use for other purposes, as one would envision in a sustainable
a reduction of the energy expenditure is desirable, if possible process.
all the way down to the thermodynamically determined All of the processes in Table 1, as well as the general case
energy difference between the starting material and product. of a chemical reaction, are comprised of several steps, which
Typical examples are the synthesis of ammonia from the can occur one after another or in parallel and build a reaction
elements, the oxidation of SO2 to SO3, and the oxidation of network. With regard to chemical kinetics, every individual
CO to CO2. In the majority of reactions, however, several step is itself a sequence of more elementary reactions. These
reaction products are possible and an acceleration of the are defined by the property that they describe the change of
reaction would lead to the favored production of the most only one chemical bond in the system. The separation of the
thermodynamically stable product. A family of such applica- complete reaction into individual steps enables the descrip-
tions is the total oxidation of hydrocarbons for energy tion of the entire process in such a way that each step can be
production or the purification of exhaust gases and of integrated into a microkinetic model with its kinetic param-
water. However, a product is usually sought which is not the eters (stoichiometry, frequency factor, activation energy).[1]
most thermodynamically stable and in fact may be less stable Unfortunately, we have only a few of these microkinetic
than the starting material. The catalyst then has the task of models.[2] In the cases where they do exist, it turns out that the
quickly activating the starting materials while slowing the mechanism that is described by a small number of individual
formation of the most thermodynamically favored products steps conceals kinetics which are complex in comparison[3] to
and thereby allowing the generation of less-stable products.
The latter should not be further activated by the catalyst, [*] Prof. Dr. R. Schlçgl
although the catalyst must be potent enough to activate the Fritz Haber Institute of the Max Planck Society
more stable starting materials. Catalysts must also often Faradayweg 4–6, 14195 Berlin (Germany)
prevent reactions to end up with the desired products, so the and
Max Planck Institute for Chemical Energy Conversion
view that catalysts always accelerate reaction rates is, there-
Stiftstrasse 34–36, 45470 Mlheim a.d. Ruhr (Germany)
fore, a misnomer. E-mail: [email protected]
This can be seen clearly in Table 1, where several Homepage: http://www.fhi-berlin.mpg.de
reactions have been listed in order of standard reaction http://www.cec.mpg.de

3466  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

the models assumed. It is possible to arrive at the same Mainly we use mechanistic conceptions for the interpre-
macrokinetic observations by starting with differing initial tive basis of kinetic models derived from macrokinetic
parameters and reaction mechanisms. Therefore, it must be principles. These data and there numerical adjustment to
assumed that there is no one single analysis of kinetic data mathematical models[2i, 15] are well-suited to draw conclusions
with respect to the underlying basic mechanism.[4] It would, about the parameters of the process, its scaling in other
however, be useful if some of the basics of physical chemistry dimensions, and the construction of chemical apparatus
were kept in mind during the selection of parameters meant within the parameter range of the observations. Considerably
to describe certain observations: the assumption of the more difficult is the use of mechanistic concepts and
production of peroxide from oxygen as a spontaneous observations of nonreacting model systems by using the
reaction without activation energy[1b] would be, for example, method of microkinetic modeling to predict catalytic behav-
worthy of consideration. ior during high productivity. This conjunction[16] has been
If there is no unambiguous relationship between kinetics used in special cases[17] with some predictive power for
and mechanism that can be used to predict the conditions for interface processes, however, a general method as used in
the favorable evolution of a reaction and the nature of the molecular catalysis is still not found in heterogeneous
active centers, it may be an exaggeration to expect[5] the catalysis.
design of a catalyst from which a favorable reaction process To proceed to the single desired product in the case of
can be obtained.[6] This includes the case when the catalyst is a general reaction with several possible products, the catalysts
formally “designed” with the help of a mechanism postulated can be designed towards a specific substrate such that they
from the simplified reaction based on formal reaction will react with exactly one component of the reaction
stoichiometry. This expectation would need to be fulfilled if network. The result is a very high selectivity and reaction
we, as is freely opined,[7] were truly able to tailor a catalyst. rate because the catalyst must only accelerate a single
Advances in theoretical chemistry may be able to make reaction step. However, a unique catalyst is needed for
inroads into this area. There are many uncertainties in the every individual step of the reaction network that does not
analysis of catalytic experiments, be they on high-perfor- proceed spontaneously. Large reaction networks require
mance catalysts or model systems, as will be discussed in the complex architectures to obtain the necessary variety of
present Review. Such uncertainties, often reworded as catalytic effects in a limited parameter range of reaction
“material gap,” “complexity gap”, or “pressure gap”, make requirements. We can also find this principle in nature.
microkinetic analysis[1a, 2f, 8] even more difficult, which is Enzymes[18] are complex molecular catalysts that cause mainly
already encumbered by chronic underdetermination of one substrate-specific reaction step in the large networks of
parameters in its mathematical models. A complete ab initio the chemistry of life. The extreme specificity of the
based model of a chemical reaction[9] with structural and enzymes[19] working together in a common reaction environ-
microkinetic components capable of describing[10] the process ment results from their complex hierarchical structures that
as a function of the chemical reaction potential at finite are composed of a limited number of elements and basic
temperatures could be a point of reference for proposing motifs. In organometallic catalysis we choose a different path
catalytic reactions. However, to reach this target, many and use catalysts with geometries, which are simple compared
hurdles must still be overcome in regard to the choice of to enzymes and attribute their specific reactivity to a very
method[11] and parameters.[12] Nevertheless, it may be large number of different ligand systems.
expected that the results obtained up until this point[10c, 13] Alternatively to substrate specificity, catalysts can also
will become an essential part[14] of catalysis science. From an work in a reaction-specific way. They allow a specific
experimental perspective, the calculation of observable sequence of reactions but are constructed in such a way that
characteristics of the resulting catalyst models would be, in they bind the starting materials more strongly than the
addition to the kinetic parameters, very helpful for building intermediate products. They then release the desired product
a connection between theory and experiment. by breaking the contact between the catalyst and reactant at
a specific point in the reaction sequence and leave the
network unfinished. This design principle requires precise
Prof. Robert Schlçgl (born Feb. 23, 1954 adjustment of the interaction between the catalyst and
Munich) studied chemistry and completed reactant so that the interaction itself adapts with the develop-
his PhD at Ludwig Maximilians University in ment of the reaction: the nonreactive starting material
Munich. After postdoctoral research at Cam- becomes strongly bound and its conversion diminishes the
bridge and Basle and a professorship at the binding ability of the catalyst. In this case, we speak of
University Frankfurt, he became a director
“adaptive catalysts” that are often used in the technology of
at the Fritz Haber Institute of the Max
Planck Society in Berlin in 1994. In 2011 he the chemical industry. Clearly defined reaction conditions
was appointed founding director of the Max ensure that the change in the catalyst–substrate interaction
Planck Institute for Chemical Energy Con- achieved through catalyst adaption quickly affects the for-
version in Mlheim a.d. Ruhr. His research mation and separation of the chemical bond between the
mainly focuses on inorganic chemistry, catalyst and reactant.
heterogeneous catalysis, nanostructures, We recognize that the effect of the catalyst always
material science for chemical energy conversion, and concepts for sustain-
requires a chemical interaction with the reactant. This
able energy supply and storage.
interaction must become more specific as more reaction

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3467
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

possibilities result from the activation of the starting materi- tions of a solvation shell would be an example of this, as is the
als. Here, the notion that a catalyst does not interact with the dissociation of an ion pair. A much-used alternative is the
reactants, even if it does not consume itself, is a misnomer. stabilization of a reactive, local structure in an isolated
This brings us to a further characteristic of catalysts. They geometric site on or in a matrix of a carrier structure.[6, 23] This
are able to carry out their interactions with the reactants could be, for example, a step on the material surface[2g, 24] or
several times and can, thus, achieve a super-stoichiometric a lattice defect[25] in the surface created either spontaneously
conversion. The effect of a catalyst is considered to be more as a result of the kinetic details of the synthesis or in a planned
potent if the super-stoichiometric conversion factor for the way by doping the matrix structure.[23d, 26] Interfaces can also
desired product per unit time increases: the catalyst is then serve as carriers for nanostructures or molecular forms of
described as being “high-performing” or “active.” To achieve active components. This common variation[27] is very chal-
an observable conversion in a chemical reaction that can be lenging in synthesis[28] and interpretation despite its concep-
considered a deviation from the chemical equilibrium of the tual simplicity, in part because a non-ambiguous differentia-
system, the catalyst itself must also exhibit a departure from tion between the effect of the carrier and the active
the chemical equilibrium of its structure. This departure can component on the desired reaction is often not possible.
be firmly ingrained in the structure of the catalyst, in which This subject is central in understanding heterogeneous
case it, as a substance, will no longer be in chemical catalysis,[27a, 30] as can be seen in the examples of gold catalysis
equilibrium. A kinetic stabilization is required if the non- or the use of polyoxometalates,[32] but will not be elaborated
equilibrium state, despite its involvement in chemical pro- upon here.
cesses, is to remain during multiple repetitions of the reaction. A further possibility is the use of a structurally dynamic
We are then dealing with a static catalyst. material. Here the fluctuations about a stabile average
Alternatively, the non-equilibrium state can always be structure cause the short and random appearances of
reached anew. In this case, we speak of chemical dynamics[21] unstable, active resonance structures. This alternative may
as the cause of the creation of active states from the structure be used frequently, although it is seldom purposefully created.
of a catalyst that remains stable on average. A realization of Much more often it is the result of chemical dynamics[33] of
such systems could be a steady state between two metal– catalyst precursors under the specific conditions[21a, 34] of their
ligand complex formation equilibria fluctuations about use (the author recognizes that the term “chemical dynamics”
a stable average structure or phase changes in bistable is used in many different ways in the literature[35] and for this
regions. Oscillating kinetic behavior[21i, 22] in macroscopic reason it will be further discussed below).
systems is a clear indication that these types of dynamic As a fundamental consequence, the search for such active
processes play a role in catalysis with molecular as well as centers cannot be successful if it is performed in the absence
solid interfaces. of the reactants. Practically speaking, we can only character-
It is expected that the performance of a catalyst can be ize such systems in situ.[36]
correlated to the extent of the deviation of its structure from Finally, repair mechanisms can be employed to reactivate
equilibrium during a catalytic cycle. However, this deviation deactivated local structures through the exchange of damaged
will affect the stability of the catalyst and also the length of elements by means of self-organization. A variation on this
time it remains effectively active. It is, therefore, not possible would be the exposure of deeper layers of an active substance
to use a high-performance catalyst over long periods of time. by separating the damaged over-layers from the underlying
Both of the desired characteristics stand in contrast to one material. Catalysts based on carbon[37] are particularly
another with the consequence that one of the most important suitable because their oxidation product is gaseous.
tasks of creating new catalysts will be formulating the We recognize a variety of different functional concepts
compromise between these important system characteristics. that we can use to synthesize catalysts. Unfortunately, the
Incorporating this compromise can be achieved in the identification of these concepts in a given and empirically
context of different architectures. In doing so, the entire found catalyst is demanding under high-performance con-
material of the catalyst does not have to be activated. It is ditions and often has not been done at all. Much more often
enough if a high-energy state, or “active center”, is formed we use the concept of empirical discovery and interpret its
out of the temporary combination of stable components from results in terms of the above-mentioned functions. Such
the system matrix. In molecular catalysis, the exchange a procedure renders knowledge-based targeted modification,
equilibrium between ligands is the characteristic process. In impossible.
heterogeneous catalysis, we know of processes between
reactants and inactive “catalysts precursors” that take place
first at critical chemical potentials. They are, therefore, 2. What We Must Investigate
responsible for the existence of “pressure and material
gaps” between experiments at low pressures and experiments Catalysis needs analysis on several scales of space and
at normal working reaction pressures. time for adequate clarification. The reason for the necessity of
A small structural instability can be utilized to gain a weak consideration on different scales is due to the large multiplier
catalytic effect over a long time period. A prototype is the use between the events on single molecules, which is of interest as
of an interface formed from the abrupt change in chemical a basis for a mechanistic description, and the chemically
bonds in a solid phase. A stabile molecule, which forms a free observable events themselves in a small laboratory reactor.
coordination site for the substrate through sporadic fluctua- There we typically observe the behavior of 1020 individual

3468 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

reaction processes and wish to describe them by understand-

ing the behavior of single molecules. Two basic challenges
appear from this large scaling factor. First, we must account
for all of the necessary steps of a reaction that include atomic
processes, making and breaking of chemical bonds, and also
energy and material transport. Second, with so many parallel
processes, a distribution of properties from relevant structures
becomes active at every scale. The uniqueness of the link
between structure and function at a purely atomic level,
therefore, becomes blurred and must be replaced by statistical
analysis. Thus, the identification of the “typical” relevant
structure for every individual step of the chemical process
becomes difficult. Although we have tools with different
resolution to analyze structures on different size and times
scales, we still suffer from considerable lack of clarity about
the conditions under which we observe a reaction. An
example of this is the resolution with which we can observe
the energetic relationships at the site of a reaction. We can
explain well[24b, 38] both experimentally and theoretically
phenomena ranging from strong interactions such as forming
covalent bonds to weaker interactions such as vibrations of
the participating molecules. Dispersive and weaker long-
range interactions are less accessible[39] and we tend to ignore
them with the argument that they do not contribute signifi-
cantly to the total energy of a reaction. Figure 1 shows
a schematic energy profile during a generalized catalytic
reaction and illustrates why this assumption may be incorrect
and that the parameter “energy intensity” of a single step is
not adequate for a proper assessment of its relevance.
Figure 1 illustrates further that the treatment of the
individual steps needed for the conversion of a chemically Figure 1. Dimensions in catalysis: A) Change in energy (E) of a catalytic
observable amount of material is split between the disciplines system with respect to the reaction coordinate (RC) for the basic
of chemical reaction technology and physical chemistry, individual steps: Tm: transport macroscopic, Tp: transport in pores or
solvation shells, Ch: chemisorption, AC: activation (dissociation), Re:
although only the description in its entirety[13b, 40] allows the
reaction, De: desorption, dissociation. The energy changes in red
analysis of a catalytic reaction. Furthermore, the Figure shows denote the impact of the reaction on active centers that are taken here
that regeneration of the active centers is essential for the to be adaptive. B) Space–time scales for a heterogeneous catalytic
catalytic function and, therefore, requires even more atten- process: blue: the molecular reaction, red: chemical dynamics on the
tion than the conversion of the reactants. Figure 1 B contains catalyst, green: transport processes of the reactants. The vertical bars
a rough overview of the space–time dimensions of a hetero- indicate often-used terms for the spacial dimensions.
geneous reaction, which may be reviewed if the behavior of
a macroscopic reactor is to be correlated with the molecular
events at the active center. the tools to realize these goals at least in small reaction
The practical success of catalysis for the preparation of networks.
structural and functional materials[41] as well as for the In Figure 2 such a small reaction network can be found for
conversion of energy carriers[42] leads us to believe that we the activation of oxygen. The network combines the fields of
have quantitatively understood the necessary basic concepts. knowledge of biological and technical energy storage with
This will allow us to at least predict which optimal possibilities that of fuel cells and of the environmentally friendly synthesis
exist for chemical reactions and which materials can be of organic oxidation products. Although the biochemical
chosen for catalyst production. If we use the enormous body reaction route seems to be sufficiently clarified,[43] we are still
of knowledge on the subject and the rules that have resulted far from a final explanation of the reaction process in
from it, we should be able to fulfill these expectations using electrochemistry.[44] This is most evident in the synthesis[45]
“chemical intuition”. of reactive hydrogen peroxide, although the possible reac-
However, if we would like to complement this with tions are limited. For the case of the activation of CO,[46]
a knowledge-based physical theory, we are at best at the a considerably more complex reaction network also offers
beginning of such an endeavor. The final goal of the “design” plenty of opportunity for further clarification. The reaction
of a catalytic process based on the knowledge of its atomic network for the formal simple activation of methane[2b,i, 47] is
details, derived for example from a theoretical description, truly complex and at this point still without final elucidation.
still lies far in the future. It is safe to say at this point that the Much of this unsatisfactory state of affairs results from the
current state of knowledge will allow us to begin developing assumption that the explanation of a reaction mechanism is

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3469
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Although the investigation of semiconductors under

working conditions is possible in many situations, the so-
called “in situ” investigation[21h, 51] of catalysts is highly
involved and thus rarely performed. Originally, these types
of investigations were started[52] to clarify why catalysts are
structurally different after use than their initial state.[53] It was
soon seen,[21a, 54] however, that this method could achieve
much more by offering insight into the reacting systems of
catalysis.
Such investigations combine an analysis of geometric or
electric structure with the simultaneous proof of catalytic
action. From this, significant methodological challenges
follow for the identification of structural characteristics
under conditions that are unsuitable for the function of the
applied method of investigation. Observing bulk or surface
structures at high temperatures and pressures of reactants
drives the analytical methods to their physical limits. The
multiscale characteristics of the behavior of catalysts are very
apparent here in the difficulties associated with measuring
kinetic data in situ gathered alongside data from correctly
dimensioned laboratory reactors. Thus, it is difficult even
Figure 2. A simple reaction network for the activation of oxygen.
Hydrogen and an unnamed electron donor are needed. Red denotes under in situ conditions to find evidence to clarify the
the formal oxidation state 1 and blue the oxidation state 2. practical function of a chemical process. The results of such
experiments have been tested with different methods of
measurement[51d, 55] by which complementary properties are
observed. It is then expected that the complete set of results
unique and independent of the conditions and catalysts used. will describe a homogeneous picture of the reaction process if
In the framework of static solid catalysts this expectation may no internal contradictions of the findings occur. The author
be justified although even there the relative abundance of prefers the view, in contrast to the literature, which recom-
reactants on the surface controlled by the reaction conditions mends integrating different in situ techniques into a single
offers multiple pathways of reaction. We will derive, that this experiment,[37e, 56] of combining several independent in situ
view of a static catalyst is only valid as an idealized boundary experiments with kinetic reactor studies and ex situ structure
case. In situations of high performance, catalysts are not static identification. A coherent description of catalyst function that
and hence a coupling of their function to the conditions used is reached collectively through such a group of experiments
will exist. This destroys the ideal expectation of a unique offers stronger evidence for the correctness of the described
mechanistic description of a catalytic reaction. The formal chemical reaction. Today, such studies are rare compared to
discrepancies in the literature may thus not be contradictions the many ex situ or “post mortem” studies documented in the
but rather snapshots of one and the same reaction scenario. literature, in which the nature of the active states must be
Catalysis is a member of the knowledge and technology inferred after the fact.
family of nanoscience.[27c] Despite its empirical maturity,[48] Unfortunately, there is a differentiation in the literature
which makes the main part of the (petro)chemical industry between “in situ” and “in operando”[36a, 51d, 57] studies, with the
possible, the conceptual strength of catalysis is, in contrast, latter having the additional requirement of having to be
not strongly developed compared to, for example, semi- performed under exactly the reaction conditions applied in
conductor technology.[49] A basic difference is that the the technical operation.[57a, 58] From the perspective of the
function of semiconductors is connected only with the control author, it would help all interested parties if a common
of their electronic structure, while the chemical and geometric definition for all such investigations could be found. This
structures remains static. In catalysis on the other hand, we definition could be: in operando investigations identify the
necessarily influence the electronic and geometric structure geometric or electronic structure of a catalyst under simulta-
during every cycle of the reaction. These functional materials neously documented production of the desired reaction
must, therefore, exhibit structural plasticity and be able to product.
change at least part of their atomic structure in a reversible The validity of this premise is illustrated by an example.
way. Here the danger of irreversible structural changes is The multielement oxide catalyst “M1” is a potent system[59]
a hazard that we observe as a loss of performance[50] for the oxidative dehydrogenation of alkanes. However, the
(deactivation). The normal description that in “catalysis” reactive phase is not the bulk structure but is rather
a previously formed active center must be reactivated as a termination phase[60] that is produced under reaction
a center of the same kind may be accurate for static centers in conditions. A critical element of the investigation is the
the limiting case of small conversion. However, this is no question of whether the termination layer is an artifact of the
longer correct for dynamic centers that are likely found much chosen method or whether it is actually the relevant phase at
more often under high-performance conditions. atmospheric pressure with a constitution that is different from

3470 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

the bulk. The corresponding in situ investigations[61] were We are striving in different ways, and with growing
carried out with NAP XPS (near-ambient pressure X-ray success,[62] to overcome this fragmentation. Dedicated confer-
photoelectron spectroscopy) at 25 Pa. ences and research networks[63] are a testament to this.
As seen in Table 2, the selectivity for the oxidative Despite this, however, a holistic conception of catalysis
dehydrogenation of ethane and propane were obtained remains an enormous challenge. This may be due to the fact
in situ and correspond well to experiments performed in that it is still difficult to amass similar know-how in the fields
a tubular reactor at normal pressure and flux. It can be of knowledge of molecular and interface-specific catalysis; in
assumed from this that the active phase has the same chemical each of the respective fields, the other field is treated in
constitution, otherwise the selectivity, which is closely linked a cursory fashion. The current Review is devoted to the
to the nature of the active centers in complex reaction understanding of the characteristics of heterogeneous catal-
networks, would be very different between experimental ysis with active interfaces which has grown considerably over
conditions. the last three decades. This is based on a “standard model” of
interface catalysis, which was developed from surface science
and supporting theory. This model still has significant gaps
Table 2: Selectivities of an M1 catalyst at different pressures.[a] with regards to transferring knowledge yielded from weakly
Selectivity [%] 25 Pa 1000 Pa or nonreactive model systems to high-performance catalysts.
These gaps occur, according to the author, from the static
Oxidation of ethane
comprehension of the catalyst during a catalytic reaction.
C2H4 98 97
CO 2 1 From this results a substantial difficulty with the utilization of
CO2 1 2 the comprehensive insight provided by the standard model for
the targeted synthesis of technical systems. This Review aims
Oxidation of propane to propose a way to bridge the gaps between heterogeneous
C3H6 51 64 model catalysis and heterogeneous high-performance catal-
CO 19 8
ysis and, for this, conceptual knowledge is taken from
CO2 29 11
molecular catalysis. This Review is not meant to be an
[a] Temperature 673 K, stoichiometric feed, flux 9000 GHSV at 1000 Pa. introduction to the different fields of catalysis, but reference is
indeed made to the fact that the combined fields of catalysis
have the same scientific roots. Here a contribution to an
3. Catalysis Science, a Basis interdisciplinary perspective on heterogeneous catalysis is
offered. Many elements which are indeed necessary for
Catalysis science today has a fragmented character. understanding catalysis as a whole remain cursory here
Catalysis as a field of knowledge offers systems and processes because of the noncomprehensive nature of this Review, for
in a rational way based on molecular concepts and inves- which the author asks the reader for forgiveness.
tigates reactions useful for the chemist in the laboratory or for
technological purposes. It is defined by extreme diversity, but
also through complexity. Next to the small section of research 4. Heterogeneous Catalysis is Systems Chemistry
that looks into this complexity, there is a larger part that
concentrates phenomenologically on the production of cata- The new and still somewhat diffuse term “systems
lysts with desirable properties without an experimentally and chemistry” denotes[64] the efforts to produce new function-
theoretically justified molecular basis. The practical success alities through self-organization from a library of different
that undoubtedly results from this approach is a proper components. In contrast to the use of materials of the highest
justification of the procedure. The great variety of catalysts, in purity in synthesis, mixtures of reactants are used in the
particular molecular catalysts, which have been and will be preparatively related systems chemistry that self-organize
found, is showcased in many issues of this Journal. The through mutual influence in such a way that novel functions[65]
current chemical industry is also largely dependent on this are produced. The minimization of the total energy (thermo-
approach. Its success explains the reticence shown by those dynamic library) or dynamic fluctuations (kinetic pantheon)
regularly employing catalysts to expend excess effort for as well as catalytic effects (“catassembler”[64a]) come into play
experimental and theoretical functional studies that cannot be as the driving force. In inorganic chemistry, this procedure has
justified with plausible clarification beforehand. A “practical” been described somewhat less to date. The production and
use derived from such studies becomes untenable, mainly preservation of dynamic, active centers in catalysts are taken
because of the amount of effort involved to overcome the as examples of such systems chemistry. Under catalytic
experimental complexities. It is one goal of the current reaction conditions, active centers are created that then
Review to show that we are indeed in possession of a concept produce products and, thereby, disappear. They are then
for escaping this “trap of complexity”. The separation into either re-formed or rejuvenated according to the procedures
“fundamentalists” and “pragmatists” in catalysis is super- discussed above. Thus, the library of components is a dynamic
imposed on a separation of the field of knowledge of catalysis one. This is immediately clear for molecular complexes used
itself that results from the nature of catalysts. Significant as catalysts. Central atoms, ligands, solvents, and reactants
disciplinary differences are found in catalysis with molecular form a library that is organized according to the laws of
systems, enzymes, and solid interfaces. association and dissociation of complexes.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3471
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

This is not so immediately clear in heterogeneous systems

with initially clearly stable catalyst materials. However, the
concept of systems chemistry provides a clue to why the
reaction conditions must always be so extreme. It is not about
enabling the activation of an organic substrate, but rather
about tuning the dynamics of the catalyst so that active
centers are produced and held free of reactants and products.
Instead of the chemistry of the desired reaction, the chemistry
of the active centers dictates the reaction conditions. This
explains, for example, the apparent contradiction of the mild
reaction conditions of the oxidation of methane in living
systems[66] with the drastic conditions that result in heteroge-
neous catalysis. The high temperatures are necessary to stop
the deposition of carbon on metallic catalysts[67] and the
decomposition of carbonates or hydroxides on oxidic cata-
lysts. Maintaining the reactive electronic structure of the
catalyst[68] requires the high temperatures[47a, 69] and not the
stability of the CH bond.[70]

5. Case Study: MgO as a Catalyst for Methane

Activation

The activation of methane to ethane (ethylene) by

selective oxidation (OCM) is a “dream reaction” for supply-
ing the chemical industry with raw materials. It is no surprise
that many attempts at achieving this are undertaken with
heterogeneous catalysis. Figure 3 shows the number of
publications and patents over time as well as a summary of
the most successful catalysts.[71] We recognize that the
Figure 3. Scientific activities in the oxidative coupling of methane
problem has not been solved sufficiently, but also that the
(OCM) as a function of time. In the lower graphic the most successful
research community has left the problem essentially unfin- catalysts are displayed in a conversion–selectivity diagram.
ished.
Notable is the observation that there seems to be
a “universal” limit to the obtainable performance, even unknown way to arrive back at the initial state. The presence
though the reaction is not thermodynamically limited. It can of alkali components in many OCM catalysts and the high
be seen further that very different solutions for catalyst reaction temperatures around 1000 K may lead to the
chemistry have been found: acidic oxides, basic oxides, assumption that there may be reactions of alkali hydroxides
mixtures, and halogen components. The assumption can be with oxygen to form alkali suboxides that serve as catalysts.
made that the cause of the “yellow line” in Figure 3 is the Unfortunately, experiments in matrices[76] and in high-tem-
complete consumption of oxygen. Thus, the course of the perature mass spectrometers[77] show that such reactions do
reaction, constrained by the explosiveness of the gas mixture, not take place practically, at least for Li, which is important in
is responsible for the “universal” limit in Figure 3, with the this case. Also no trace of the presence of LiO was found in
local chemistry at the active center only being indirectly the relevant model experiments using EPR.[69b] However, this
related through its low selectivity. may be different when methane is introduced into the system.
The reaction was studied in detail using the catalyst In any case, the possibility of a continuous catalytic cycle with
system Li-MgO and from it a general reaction pathway[72] was LiO is doubtful, even more so if the simultaneous presence of
deduced. However, discussions of this mechanism have led to water and CO2 in the reaction mixture is considered. It can be
controversy of late.[47a, 69b] Initiating the debate was the found for MgO in extensive theoretical treatments[23d, 26, 78]
observation, seen already earlier, that MgO[73] itself also that if this aspect is disregarded, then the substitution of
exhibits appreciable activity in the OCM. It was shown alkaline earth atoms by Li causes a substantial reduction in
further that the Li component is leached out of the the activation energy of the OCM and is in a sense similar to
catalyst[50b, 74] and then acts as a structural modifier.[68a, 75] electronic doping; this is valid for the doping of the (100)
Independent of this, the basic conception of the mechanism surface as well as for the edge of a step.
according to Lunsford should be used as a guide. He Staying with the basic idea of Lunsfords concept that the
postulates that the critical reaction is the activation of activation of oxygen is a difficult step, there is only one
oxygen into an atomic radical state (M-O*). This species possibility if the alkali hypothesis is omitted. The electrons
can activate methane and convert it into a methyl radical and necessary for the production of the reduced oxygen species
the then deactivated M-OH can react with oxygen in an must come directly from the methane. The participation of

3472 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

MgO as an electron donor in a pure or defect form[69b, 78] can Evidence for this hypothesis was gathered with EPR
also be discounted. We learn from this about the functionality spectroscopy: the hyperoxide radical could only be found on
of the catalyst that comes very close to the original conception an activated MgO if methane and oxygen were present. If the
of catalysis in general: a catalyst is a material that, through its catalyst were to activate oxygen on its own, perhaps through F
presence, affects a reaction of two components that would centers, then the hyperoxide radical would also have to arise
otherwise not react. The MgO acts as a “marriage broker” in without the presence of methane. This experiment,[80] docu-
the Reaction (1) without itself supplying the electrons neces- mented in Figure 4, also supports the hypothesis that a high
sary for the activation of oxygen.

CH4 þ O2 ! CH3 * þ OOH* ð1Þ

This activation reaction leads to a series of subsequent

steps that we will now consider. First of all, there is the
process[79] of the initial step. Necessary for this is the
adsorption of methane and oxygen on MgO. This can
happen in an advantageous way for methane because
Mg2+O2 ion pairs exist on the surface that are not fully
coordinatively saturated because of the incomplete coordi-
nation at the interface. They polarize the symmetric methane
molecule and allow adsorption through a C-Mg and H-O
interaction. For an isolated MgO*+ radical cation,[47a] this
interaction would lead spontaneously to the production of the Figure 4. EPR spectra (X band) of MgO after heating in a vacuum at
methyl radical. The altered Mg-O binding relationship will 1073 K. The measurement was performed at 77 K. For spectrum (a),
only oxygen (50 mbar) was introduced, whereas for spectrum (b),
occur preferentially on the (100) plane of a crystal surface.
oxygen and methane were introduced. Three different axial anisotropic
This is not enough, however, for a dissociation of the methyl environments can be seen for the hyperoxide radical.
radical because the rather ionic state Mg2+(CH3) provides
a substantial stabilization.
The catalytic effect of the MgO step is found in the fact temperature is not required for the activation of methane.
that an oxygen molecule can bind to the formal ion pair. The Furthermore, the complexity of the spectrum shows that there
negative charge moves to the oxygen and the methyl species, should be several local environments for the hyperoxide
is thereby set free and a hyperoxide radical anion is created. radical and that other surface structures are also reactive
This can then accept the proton from the MgO step edge and besides these steps. However, the same reaction pathways
also desorb into the gas phase at the high reaction temper- may not be followed by all adsorbates, because their local
ature of 1000 K, at which point the following reactions make environment is different.
the reaction pathway difficult to interpret. Shown as Equa- After this heterogeneous catalytic reaction there is a net-
tions (2)–(4), the step edge of the MgO (MgO’) causes the work of subsequent reactions that lead to the observed
following reaction without the participation of electrons from distribution of the products CO2, CO, C2H4, C2H6, and H2O. A
the catalyst: very simple, possible map of the reaction pathways[2b] is shown
in Figure 5. Further radicals appear here and it is not clear[1b]
O2 þ CH4 þ MgO0 ! CH3 * þ MgO-H þ OO* ð2Þ whether all reactions take place in the gas phase, perhaps
through interactions with the entire surface of the MgO. The
OO* þ MgO-H ! * OOH þ MgO0 ð3Þ complexity of the reaction progression becomes clear and is
shown compactly in summed notation in Equation (8):
* OOH þ CH4 ! CH3 * þ 2 OH* ð4Þ
2 CH3 * ! C2 H6 ð5Þ
The appearance of the hyperoxide in Equation (2)
C2 H6 þ OH* ! C2 H5 * þ H2 O ð6Þ
corresponds to the reacting species in the mechanism of
Lunsford.[72] The catalyst[79] makes it possible for all of the C2 H5 * þ OH* ! C2 H4 þ H2 O ð7Þ
necessary elementary reactions[1b] making up the steps (2)–(4)
to be temporally decoupled and to proceed with small Sum ð2Þ  ð7Þ : O2 þ 2 CH4 ! C2 H4 þ 2 H2 O ð8Þ
activation barriers. This is because the fluctuations of the
electronic structure at 1000 K can ensure that the appropriate For the dehydrogenation of the initially produced meth-
local electron configuration[47a] will occur before the electron ane, there are several possibilities in addition to reactions (6)
transfer takes place. The hyperoxide radical, or its protonated and (7), which are shown here only for formal reasons. This is
form, can eventually remain on the surface until, as shown in evident in the reaction network in Figure 5.
step (4), another methane molecule is activated which should To verify the effect of the catalyst as a facilitator in the
then lead to the fast decomposition of the peroxide into gas- targeted reaction (8) five samples of highly pure MgO were
phase OH radicals. produced[81] that differed from each other only in their

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3473
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

dition for catalytic activity, it is not

a sufficient condition. The active centers
relate to surface defects of MgO, the
densities of which differ with the prep-
aration method.
The method of selective adsorption
with spectroscopic detection was
applied to further characterize these
defects. Fine details of the local geo-
metric structure can be observed
through analysis of the vibration spectra
of adsorbed CO an MgO. After 30 years
of detailed study[84] we have amassed
a catalogue of vibration frequencies that
correspond to specific structures, includ-
ing a value (2147 cm1) for the mono-
atomic step edge of the (100) surface.
We deduced this classification from the
observation that the adsorption of CO
at 77 K on the more weakly coordinated
terrace sites (2156 cm1) and multilayer
steps (2170 cm1) are not observed if the
sample is thoroughly cleaned of
stable[85] O-H groups before adsorption.
The more stable bond of CO on a single
step (Figure 6 A) results from its two-
fold coordination (Figure 6 B). The
number of surface defects can be infer-
red from the quantitative analysis of the
intensity of the absorption bands caused
by this specific CO adsorption. This
number is characteristic of each syn-
thesis product of MgO (extrinsic) and
not characteristic of MgO as a material
(intrinsic). Figure 6 C shows a HR-TEM
image of MgO nano-crystals with
atomic resolution. The single and multi-
step arrangement in the cubic structure
can clearly be seen and also that the
single step edges are not the most
common type of surface defect. Further-
Figure 5. Simplified reaction network for the activation of methane over MgO. The double
more, it can be clearly seen that the
appearance of “CH3” is for clarity and has no mechanistic purpose.
catalytically relevant centers are few in
comparison to the total number of
centers that are present on the surface
geometrical structure. Trace amounts of foreign elements and in the bulk of the nanostructured particle. Thus, the
were detected in all samples, but no relation was found identification of active centers becomes the problem of very
between their presence in ppm amounts and the observed high analytical sensitivity; active centers are very rare indeed!
reactivity. In accordance with the idea[3a, 24b, 82] that active From the coadsorption of CO and methane and quanti-
centers are characterized mainly through local electronic tative analysis of the resulting spectra with infrared spectros-
conditions, the differing morphologies lead to different copy, verification was obtained that monoatomic step edges
kinetic rates for the activation of methane. The structural play an important role in reaction processes at least at the
sensitivity[83] of a catalytic reaction observed here and else- beginning of a catalytic reaction during the first 100 h.
where, illustrates the particular bonding arrangement at the Figure 6 B shows an IR spectrum of a coadsorption experi-
active centers. From this observation it was realized further ment together with a spectrum for pure methane on MgO. CO
that the terraces of the geometric surface of MgO couldnt be blocks the monoatomic step edges selectively under the
the location of catalysis because there was no correlation chosen conditions and significantly reduces the adsorption of
between the amount of terraced surface area and activity. This methane. The number of adsorption sites can be determined
shows that although adsorption is indeed a necessary con- from the difference in the spectrum intensities of adsorbed

3474 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

A structure–function correlation has been derived

(Figure 7) from the combination of targeted synthesis, inves-
tigation of catalytic activity free of macroscopic transport
barriers, and the investigation of a specific surface defect. We
see that the catalytic function of the OCM cannot be
correlated to the substance MgO itself but rather to a specific
defect structure. Through the combination of information
from Figure 6 B and C, we were able to identify the single step
edge as a relevant structure.

Figure 7. Correlation of the consumption rate of methane as well as

production rates of the C2 target products with the area of the
adsorption bands of the IR spectra of CO at 2147 cm1 resulting from
the deconvolution of the complex band forms. Overlap with absorption
from other arrangements of CO on MgO defects is responsible for
most of the overall error.

It is noteworthy that the selectivity of the reaction to the

coupling products takes the same course as the activity. It is
uncommon that the selectivity of an oxidation reaction
increases with increasing conversion, and this can be seen as
a strong indication that the reaction pathway is indeed
described by Equations (2)–(4). The catalyst activates oxygen
in a way that is not independent of methane because it
enables the immediate transfer of electrons from the methane
to the oxygen. The mechanistic coupling of the surface
coverage of both reaction products near the reaction sites
Figure 6. A) Model of a step in the (100) plane of MgO. The oxygen does not mean that, in a macrokinetic observation, we should
ions (red) are shown with a realistic size compared to the Mg2+ ions expect equal formal reaction orders or partial pressure
(gray). A polar (111) step is displayed in the inset as a comparison. dependencies. Relevant is the presence of species at the
B) IR transmission spectrum of MgO after adsorption of methane surface, which is related to the partial pressures by the
(solid line) and after coadsorption of methane and CO (dashed line; sticking coefficients under the reaction conditions. The
measurement temperature 77 K, adsorption pressure 5 mBar). The
critical relevance of sticking coefficients will be discussed
schemes reflect the bonding arrangement of the adsorbates. C) Aberra-
tion-corrected TEM image of a MgO nanoparticle. further below.
The deviation from the correlation of the sample (HT)
toward lower values and of the sample (C) toward higher
values probably indicates the existence of rough terraces in
methane. This is then compared with the catalytic activity in (HT) and especially smooth terraces in (C). Rough surfaces
a structure–function correlation. A comparison of the IR also result during deactivation of the sample. After approx-
spectra of adsorbed and free methane demonstrates the imately 250 h, the reaction stabilizes at a low level, although
polarizing effect of MgO. Finally, a red-shift of 14 cm1 is every system is different, and at this point tends to burn more
observed for the C-H stretching mode and the appearance of methane. The corrosion caused by the water[86] that develops
a symmetry-forbidden symmetric C-H stretching mode at during the reaction turns the samples into Mg(OH)2, which
2897 cm1 can be recognized. then becomes dehydrated. The (100) steps are depleted

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3475
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

through this process and surfaces[88] with higher indexed

steps[85, 87] are produced and can be described with models
using (111) steps. Such steps are polar and, therefore,
saturated with OH groups.[85] From their topology, shown in
Figure 6 C, it can be seen that these terminations contain
exposed Mg centers while their environment is heavily
screened by the OH groups. Therefore, it can be concluded
that a reaction on such rough surfaces will proceed in another
way, and we assume that different reaction pathways will
result depending on local surface geometry. This also supports
the evidence for the differing local environments of the
hyperoxide radical (Figure 4). It can not be ruled out that the
reorganization of the MgO ion results in accumulation of
contaminants at the surface that through activation of oxygen
allows redox chemistry on Mg(X)O centers. Despite intensive
searches, there has been no evidence for this.

6. The Nature of an Active Center

The long-held and chemically proven notion[89] that oxide

surfaces are covered by dissociated water and are, therefore, Figure 8. Intermediate steps for the activation of methane on MgO:
protected from the chemisorption of other molecules onto the (1) on a free (100) step, (2) on a (111) O-terminated step. The
oxide surface was clearly confirmed for MgO through in situ hydroxylation is not shown.
spectroscopy.[90] This was then further corroborated theoret-
ically[85] in combination with model experiments.[91] From
these studies we conclude that the reactive centers of MgO for permanently bound into the material matrix represents such
methane activation are blocked in the presence of moisture. a general possibility.
This is another reason for the necessary high reaction We return now to the discussion of the case study MgO.
temperature that, again, has nothing to do with the stability Figure 8 shows the significant structures of the reaction
of the CH bond in methane. There is fierce competition pathways over steps and isolated centers. It can be seen that
between the dissociative adsorption of methane at an active both pathways lead to the desired product CH3*. The
center and the corresponding reaction with water. This subsequent path to CH3OO* is easier for the isolated center
interaction is expressed in Equations (9) and (10): that, according to the reaction network (Figure 5), can easily
lead to over-oxidation of methane. This may be the explan-
-O-Mg-O-Mg- þ H2 O ! -O-Mg ðOHÞ -OðHÞþ -Mg- ð9Þ ation why the initial activity of MgO and of the analogous
CaO[26] occurs through the (100) monoatomic step edges and
-O-Mg-O-Mg- þ CH4 ! -O-Mg ðCH3 Þ -OðHÞþ -Mg- ð10Þ through smooth inactive (100) terraces. As active centers, the
steps are too unstable in moist reaction environments. The
The reaction temperature must be chosen in such a way chemical dynamics of the MgO converts them into rough
that all possible active centers are free of dissociated water hydroxylated terminations on which only a common reaction
but must remain low enough that the water from the reaction of O2 and CH4 on a single Mg center is possible in the absence
does not immediately convert MgO into Mg(OH)2. This of heteroatomic doping.
works poorly in the case of pure MgO and explains the We also see that the determination of the size of the active
notorious instability of this catalyst in OCM. It has been center is problematic. In both reaction pathways there are
suggested that this clarification applies equally to many other different numbers of atoms involved, although always more
oxide systems (see Figure 3), especially if they contain alkali than one. Here, long-range influences were still ignored
and alkaline earth components. despite their mechanistic importance, for example the effect
Monoatomic steps represent outstanding local geometries of the polarization of methane on the Lewis acid–base pair
for reactions where the dissociation of a stable molecule is Mg-O.
important. Centers with two free coordination sites are As a consequence of the possibility of remotely manip-
adjacent to centers with one free coordination site and show ulating the active center, which was demonstrated through
a slightly larger separation over the diagonal of the step a model experiment,[93] a concept for the promotion[80c] of the
profile than in the more-stable terrace sites. This concept of reaction was developed. It is advantageous (Figure 8 (2)) to
active centers[3a, 50g, 92] is fundamental in the catalysis of activate the oxygen atom at a center located adjacent to the
dissociative reactions. However, it should be stressed here adsorption center for methane to impede the production of
that other possibilities for the catalysis of dissociative the CH3OO* radical. If this is not geometrically possible on
reactions exist, also on surfaces, which are not redox active. MgO in its stationary state, then a foreign atom must help.
Doping[93] of planar boundary layers with foreign atoms This would be a redox-active cation[23d, 26, 30d, 94] such as Fe, Mn,

3476 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

or Co in the MgO matrix. However, it is quickly evident that light on empirical principles of synthetic chemistry will not be
this is problematic because under the reaction conditions the elucidated here. In the following, the focus will instead remain
initially isolated cations aggregate to nanoparticles of the on identifying and understanding the causes for the still-
promoter oxide that, at high temperatures, are excellent deficient state of knowledge in heterogeneous catalysis. The
oxidation catalysts for converting methane into CO2. This present discussion is not aimed at creating a model that can be
undesired consequence of chemical dynamics can be pre- understood well in textbooks[100] and original research
vented with the following concept. A gold atom on the surface papers.[97a, 101] We are more interested in the possibilities for
acts as the targeted anchor for oxygen, exactly as in other further development that can be obtained through consider-
applications where the perimeter between the gold and its ation of past advancement.
carrier promote the production of activated oxy- The roots of the standard model can be found in the ideas
gen.[6, 23d, 26, 30d, 94, 95] A single gold atom that is activated[30d, 93] of Langmuir and Taylor. Both studied typical model reactions,
into a charged state by a transition-metal atom (or a small such as the oxidation of CO and H2 on Pt or the hydro-
cluster of atoms) buried under the surface or in the bulk genation of CO on Ni, and attempted to formulate a “theory
avoids direct contact between a redox promoter and organic of catalysis”.[102] Taylor observed an enormous specificity in
molecules and, thereby, prevents their total oxidation. catalytic effects. The smallest addition of foreign materials or
Completely “unexpectedly”, the reference experi- a pretreatment changed the properties of “hydrogenater
ments[80c] gave the result that the monoatomic steps are nickel”. As described in the case study of MgO (Figure 6), he
actually active centers: Doping MgO with gold atoms alone was unable to detect any changes with X-ray diffraction,[51k, 103]
caused the activity to drop almost to zero because the gold a typical bulk-sensitive analytical method. He concluded from
atoms accumulated almost exclusively on the step edges and this that the vast majority of atoms in catalysts have the same
made these inaccessible to the reactants. This is an example of geometric arrangement. This should also be the case for the
a synthetic concept in heterogeneous catalysis. If the reac- sample surface. However, because the catalytic action of the
tivity of a dissociative reaction is to be reduced, this can be samples was clearly different, a small minority of atoms must
achieved with atoms that collect preferentially on step edges be essential that he could not detect with his analytical
(gold, lead, tin, sulfur, etc.) and can be carefully controlled by method. This minority, which he called “aristocratic
adding only small amounts at a time. Among other uses, this atoms”,[104] was introduced as active centers. For these, he
technique is also employed in the catalysis of selective developed the idea of coordinative undersaturation and
hydrogenation. postulated that atoms on crystal defects have fewer neighbors
than on average in the crystal and, therefore, can be reactive.
The catalytic reaction presupposes adsorption, although this
7. A Standard Model for Heterogeneous Catalysis does not necessarily lead to a reaction on its own, a result we
were able to see in the case study on methane activation. The
In the last 100 years, vast strides have been made in the degree to which this description of active centers is accurate
general understanding of the function of heterogeneous can be seen in Figure 6 C.
catalysts. From measurements of the temporal laws governing In metallic systems, in addition to ordered surface regions
catalytic reactions we have proceeded to a quantum mechan- with translational symmetry, there are also regions that are
ical based molecular understanding of the elementary steps rough and jagged, as suggested by Taylor[102] and illustrated in
and a microkinetic description that is able to explain macro- Figure 9. Together with Taylors model, copper nanoparticles
kinietic observations. With this we have clarified the “mech- that are active in methanol synthesis can also be seen in high-
anism” of a catalytic reaction.[17] Unfortunately, this approach resolution aberration-corrected TEM images.[56e, 105]
is only successful for a limited number of reactions and,
importantly, we cannot yet explain the reactions crucial to the CO2 þ 3 H2 ! CH3 OH þ H2 O ð11Þ
transformation of our energy systems toward increased
sustainability. Although we are apparently in possession of CO þ 2 H2 ! CH3 OH ð12Þ
this fundamental understanding in a quantitative form and
also have the “standard model” of heterogeneous cataly- A rich variation of defects in the bulk of the nanoparticles
sis,[22d, 24a, 97] we are not able to treat more than a small number ensures that the surface also contains a high density of defects.
of specific cases. With the word “treat” we understand the Strain and foreign atoms in the defects stabilize this high-
quantitative explanation of the reaction pathway under high- energy state in “methanol copper” in such a way that it is not
performance conditions, the development of a resilient struc- lost during the reaction, even under drastic conditions. This
ture–function relationship, and the prediction of possible was established by a profile analysis from neutron diffraction
improvements[41e, 98] with experimental verification. This experiments[106] under reaction conditions. The diffraction
would characterize the ability of a “mature” scientific field profile of such defect-rich systems can be measurably[107]
and technology to be equal to the task of meeting future different[51k, 103] from profiles on defect-free samples and can
challenges with an adequate set of tools. be used for the quantitative and temporally resolved obser-
It has been suggested that certain areas of homogeneous vation of defect reactivity.
catalysis have already reached this point; however, whether A quantitative confirmation[56e] of the Taylor hypothesis
these most impressive results[41a, 99] are an indication of was achieved with these data. The results are given in
a fundamental understanding or whether they rather shine Figure 10. A family of samples of copper nanoparticles

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3477
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Figure 9. Metal nanoparticles show partially exposed metal atoms.

A) Model of active centers[102] from Taylor. B–F) Aberration-corrected
TEM images of nanoparticles of copper. (B) shows a bare surface. The
blurring of the termination atoms is caused by stimulation (motion)
during the TEM measurement. C–F) Magnification series on a nano-
particle (region shown in green), showing different varieties of defects
in a single particle.

Figure 10. Structure–function relationship for Cu nanoparticles. A) Cat-

alytic activity at 483 K (blue) and 523 K (red) in a stationary state for
supported on ZnO was produced.[105a, 109] The catalytic behav- a stoichiometric reaction mixture at 60 bar pressure as [mass meth-
ior in methanol synthesis correlated poorly with the “active anol/mass catalyst]  Cu surface area in arbitrary units. If the Cu
surface,” as determined with N2O adsorption,[110] but very well surface were the controlling value, the result would be parallel to the
x-axis. B) HR-TEM image of the interior of a Cu nanoparticle showing
with the stacking defect density in the bulk of the 5–10 nm
the dislocation in the stacking sequence of the atoms. The red line
particles. The terminations of twin boundaries and dislocation shows the dislocation of a row of atoms relative to the nominal
boundaries on the surface[56e] causes the creation of stabile position through the stacking sequence. C) Stacking errors effect
steps that are apparently an important requirement for a change in the surface termination of Cu nanoparticles. Three twin
coppers role as an active center in the synthesis of methanol boundaries (yellow) change in distinct ways (red: (100), blue: (110),
from CO2 that follows Reaction (11). green: (111)). Taken from Ref. [56e].
The idea that high-energy sites act as active centers for
catalytic reactions is generally accepted today. However,
there is still a conceptual problem. There are many forms of only several examples of “defects”. Their total number is
high-energy sites at an interface: point and extended defects small compared to the total number of atoms arranged with
in the bulk and on the surface, contamination, segregation, translational symmetry. As the defects have various structural
and morphological defects caused by strain and stress in the and electronic local properties (see the exact position of the
active material or induced by the support, as well as changes atoms at defect sites in Figures 6 C, 9, and 10 C) it is an
in the electronic structure at the boundaries of active enormous challenge to identify and then classify them all. We
materials with and without the influence of the support are also need a method of differentiation because not all types of

3478 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

defects are reactive. The possibilities of advanced catalyst ods to determine the upper limit of the number of active
characterization necessary for such classification are, in centers of a catalyst.
principle, available today. However, they are seldom used A notable exception to this statement is the determination
because the costs of such studies, some of which are illustrated of the number of active centers in the heterogeneous catalytic
here, are still high and their priority is thus lower than that of metathesis reaction:
the empirical search for catalysts.
We should keep in mind that a general quantitative 2 C3 H6 ! C2 H4 þ C4 H8 ð13Þ
determination of active centers in heterogeneous catalysts is
still problematic today. That was also the case in Taylors time The catalyst is a highly dispersed MoOx system that in this
and has not improved much since then. However, we are now case[116] was dispersed on SBA15. Reaction (13) is made
better equipped to estimate the dimensions and challenge of possible because a prefabricated Mo-CH2 carbene enters in
the problem through the available analytical methods and our the reaction cycle. By capturing the carbene with isotope-
knowledge of elementary reactions.[22d, 97a] A motivation to labeled ethene the number of carbene molecules present
proceed further along this path can be seen in the success of under the reaction conditions can be exactly determined to be
syntheses based on models of high-energy centers. These 1.5 % of the available Mo centers. In further experiments, the
studies[6, 22d, 24a, 111] have contributed significantly to the estab- generation of the carbene could be verified through a preced-
lishment of the concept of active centers as the basis for the ing redox reaction of propene with Mo centers that were
standard model, although there is little experimental evidence initially hexavalent, and tetravalent after the reaction. The
for them during catalytic action. geometry of the Mo-carbene is not optimally suited for the
The practical difficulty of an unambiguous determination reaction, as can be concluded from a comparison of the TOF
of the number of active centers during a catalytic reaction also of 0.15 s1 justified for this case with the reference molecular
clarifies the paradox of why we cannot adequately specify the carbene (TOF of 0.9 s1). In a following study[117] a combina-
most important characteristic of a catalyst: its activity. tion of in situ NEXAFS and simulation of the theoretical
Although the number of molecules converted per unit time spectrum found that the number of active centers is so small
can indeed be easily ascertained, this can only be done only because they are highly geometrically frustrated with
relative to the observable quantities mass, volume, or geo- respect to the thermodynamically stable Mo(O4) geometry.
metric surface area of the catalyst used. The relevant This analysis should be one of the most exact experimental
reference value would be the number of active centers in determinations of the number and nature of active centers. It
the system, but this cannot be determined. Therefore, the was also enabled by the chance occurrence of several
elegant concept of the “turnover frequency (TOF)”, which favorable circumstances, of which the low necessary temper-
leads us to believe that we can in fact measure the specific ature was especially significant. Still unexplained are the
activity of a system, is only an idealization with many sources dynamics of the generation of these centers and how they lead
of error resulting from the necessary assumptions of its to the target reaction (13), which is apparently impeded
derivation. The “inventor” of this concept, M. Boudart, compared to purely molecular catalysis.
commented upon this problem in his publications[101d, 112] but it
did not stop him from making use of his idea with “suitable”
approximations. 8. Quantification of the Standard Model
The seemingly understood concept of the TOF is often
found in the derivation of the kinetics of catalytic reactions, in Despite the difficulty in ascertaining the number of active
the theory of kinetics, and in the comparison of the centers, it is still possible to describe quantitatively the
effectiveness of often very different catalysts. For this kinetics of specific heterogeneous catalysts. At the heart of
reason we will now discuss several approximations. Unfortu- the concept[24a, 92a, 96a, 111b, 118] lies the idea that a heterogeneous
nately, a vast number of publications, which use TOF values reaction cannot function without the adsorption of the
do not state which approximation was used and, therefore, reactants. Therefore, adsorption and its inverse process
caution must be exercised when comparing reported absolute desorption are central processes in the kinetics of a heteroge-
values. An approximation that is often used for unsupported neous catalytic reaction. We have I. Langmuir to thank for the
metal systems is to use the number of surface atoms as the quantitative formulation of this concept. In 1922 he published
active site count, which results from the surface geometry his detailed “Theory of Catalysis”[119] that makes up the
(TSA for total surface area). In the case of supported metal central element of the standard model as the “Langmuir–
systems, the reference for the estimation of the number of Hinselwood Mechanism (LHM)”. Behind the double name
surface atoms is an area that is measured by chemisorption[113] there lies a discovery, already known in the 1920s, that the
or electroxidation[114] of a probe molecule, often H2 or CO same reaction (H2 + O2 over platinum) leads to different
(ASA for active surface area). In only a few cases do we have kinetics at different pressures and that the catalyst has
measurements in which the probe molecule is a reactant and a “memory” of previous treatments.[120] The missing pieces
conditions for the adsorption measurements can be chosen to the universal microkinetics that would later be designated
close to reaction conditions without actually triggering the as the “pressure gap” and the “material gap” were known
reaction.[61, 115] This is referred to as an RSA (reactive surface long before the identification of their causes: “However this
area) characterization. Despite the plurality of methods and may be, the surface on which reaction takes place is not the
further refinements, today we only have approximate meth- same at normal pressures as that at the low pressures of the

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3479
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Langmuir experiments.”[120] It is also noteworthy that Lang- The inclusion of hydrogen and oxygen in metals according
muir states at the end of his publication on the theory of to Equations (14) and (15) occurs either during the synthesis
catalysis: “At low temperatures (300 K to 600 K) rather erratic (nonstoichiometric reduction) or results from the activation
results are obtained for the reaction velocity with mixtures of of reactants. It is noted that these types of solid solutions with
hydrogen and oxygen, for the velocity depends upon the main group elements are also of importance in other areas of
previous treatment of the platinum.”[119] This observation did applications of metals.[124]
not stop him from assuming a constant number of active The standard model supplies the kinetic approach of the
centers and a relative independence of their function. LHM with a molecular basis by normatively answering the
The silver catalyst for the oxidation of methanol to question of how to count active centers. The single-crystal
formaldehyde[122] (BASF process) and platinum in the oxida- approach[97a, 127] states that the function of the active catalyst
tion of ammonia to nitrogen[123] (Ostwald process) illustrate can be described by a single crystal of a suitable material and
drastic examples of the instability of a catalyst surface. The correct surface orientation, including its steps and boundary
radical change of the bulk of the metal at a 523 K working atoms.[24a, 92a] By using the concept of the TOF,[101d] the
temperature, that is far below the melting point, can be clearly absolute number of the centers is no longer necessary. The
seen in Figure 11. The underlying processes are not confined single crystal and its translationally symmetric structure allow
to the reacting interface, but rather also span the bulk of the the application of the tools of surface physics before and after
the test[24a] in a catalytic reaction. Adsorption and reaction
experiments can be quantitatively described[24b, 38] and the
structural sensitivity of adsorption and reaction experimen-
tally and theoretically investigated. With this, the elementary
steps can be described with observable kinetic constants.
A resulting kinetic model then describes the process of
a catalytic reaction with a theoretical ab initio derived
mechanism[4a, 16, 128] and quantitative, uniquely determined
kinetic constants. For the classical case of the synthesis of
ammonia,[128] minor corrections for the original kinetic data
determined on single crystals resulted from detailed experi-
ments by several groups, but as it turned out, the corrections
had no significant influence on the outcome of the simu-
lations.[129] This was because the changes largely compensated
each other. From this work it becomes apparent how difficult
it is to bring kinetics and mechanisms into agreement; there
are often only ambiguous relationships[4b] even when stringent
mechanistic predictions prove pertinent in several cases with
limited parameter sets.[17] This conceptual approach of linking
surface science with theory was first performed for the
synthesis of ammonia from the elements. It ended the debate
Figure 11. Catalytic etching in mesoscopic dimensions (in SEM). A– on the reaction mechanism on this one of the most valuable
C) Silver in oxidation catalysis. The particles from (A) are completely reactions mankind has developed,[130] and marks the end of
sintered with one another after 100 h operation during the epoxidation long efforts in physical chemistry[131] to reach an understand-
of ethene at 523 K. Picture (B) shows the generation of crystal facets ing and clarification of the catalysis of this reaction.[22d]
by the transport of bulk atoms: the small bright objects are particles
A simplified schematic description of a heterogeneous
of silicon contamination originally dissolved in the bulk that segregate
as oxide. C) A single-crystalline sphere illustrating that catalytic etching reaction is given in Figure 12. It is apparent that the reactant
is structure-sensitive and does not attack the (111) surface. The bright must initially move to the region near the surface (not treated
objects are relocated silver metal. D) Platinum, also after methanol in the model) before being adsorbed there. This mainly
oxidation. See Ref. [126]. Noticeable are the differing orientations of spontaneous reaction (1) from Figure 12 brings the molecule
the grains that can be identified through the direction of the visible in contact with the surface through the formation of
(111) facets.
a chemical bond (chemisorption). The process must be
differentiated from adsorption by dispersive interaction,
which is denoted as unspecific adsorption or as physisorption.
metals and set free dissolved heteratoms. The holes in the Often the molecule needs to be further activated, which
sample arise from gas eruptions, whose reactants are trans- necessitates a stronger interaction with the catalyst to
ported by bulk chemistry to grain boundaries. surmount the activation barrier (see schematic Figure 1 and
for example Ref. [22d]). The position of the molecule changes
ðAgÞn þ H2 ! Agn H ð14Þ in relation to the surface, which results, for example, in a side-
on interaction ((2) in Figure 12). This process of reorienta-
ðAgÞn þ O2 ! Agn O ð15Þ tion[2e, 14a, 34b] requires energy. In addition, islands of the
reactants can form if strong bonds are present. Such strong
2 Agn H þ Agn O ! 3 ðAgÞn þ H2 O ð16Þ adsorbates block the surface for further adsorption and lateral

3480 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

low pressures. We can see further that the molecules and

atoms will have to move on the surface and that for this
reason the available free space on the surface is a decisive
quantity for the kinetics of the reaction. The involvement of
two reactants clearly makes the entire situation more
complicated. As soon as we have familiarized ourselves with
the fundamentals of the quantitative model we will be able to
discuss the questions central to catalyst design: How does the
nature of the surface influence the availability of free sites
under a given set of reaction conditions?
A classical example for the process shown in Figure 12 is
the decomposition of NO on a stepped Ru (0001) surface:

2 NO ! N2 þ O2 ð17Þ

Figure 13 shows the result. A monatomic step as local-

ization for the dissociation of NO molecules is evident. The
dissociation on the step leads to a distribution obeying
a diffusion law for nitrogen atoms (gray squares) and to the
nucleation of an oxide phase removed from the step (dark
cluster). The conceptual similarity with Figure 12 can be seen.
Directly underneath some of the step edge atoms are metal

Figure 12. Simplified representation of the steps of a heterogeneous

reaction according to the standard model. Steps (1)–(4) are shown in
a top view of an ordered surface with two types of atoms and
a monatomic step. They represent: (1) molecular adsorption, (2)
island formation, (3) diffusion to the active center, and (4) dissocia-
tion. Steps (5)–(7) are shown as cross-section through the surface.
The reactant is a diatomic molecule (blue/green). They represent:
Reaction (A) desorption, reaction (B) desorption of a product and
generation of a surface bond of the other product, (C) desorption of
a product and the generation of a subsurface bond with the other
product.

Figure 13. STM image of the decomposition of 0.3 L NO on (0001) Ru

transport.[132] Finally, the molecule must arrive at an active at 300 K after 30 min reaction time. The dark line depicts a top view of
center to achieve dissociation (the step in Figure 12). a monatomic step. From Ref. [130c].
At the active center there must be ample room for both
products of the dissociation (3). Two activated atoms must
come together in order that a reaction can proceed to atoms (type I) while underneath other steps the underlying
a product (4). Then there are essentially three possibilities for atoms are somewhat more distant. We thus identify two
the resulting reaction, of which only reaction (A) to the “types” of atomic steps even in this very simple geometric
products (5) is given by the standard model. Products are situation. For this reason the oxygen atoms form a row of
formed which leave the surface through desorption. The “oxides” on the type I steps which become contaminated (at
standard model also provides for the reverse reaction to the 200 K), while the oxygen does not adsorb on the type II steps
starting materials and to their desorption, and with this the so that they remain active.
chemical equilibrium can be established. The active center In the standard model we assume a materially unchanged
and the entire surface are not changed and are, therefore, catalyst. In many cases, however, an activated molecule can
ready to host another reaction cycle. react irreversibly with the catalyst and modify it. The reaction
It is apparent that many steps are necessary to complete pathways (B) and (C) in Figure 12 indicate that deposit layers
a catalytic reaction cycle even in this simplified picture that is can be formed. Such processes often begin at high-energy
suitable for describing model experiments on single crystals at centers and a slight deposit layer (approximately 5 % of

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3481
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

a monolayer) can completely stop the reaction because The simplification about universal active sites does not
steps (3) and (4) in Figure 12 are prevented. Following apply to many selective reactions. Either the active centers
reaction pathway (C) from Figure 12, an activated atom can change their chemical reactivity during the reaction (adaptive
also penetrate into the bulk of the catalyst instead of centers) or there is a set of centers with differing functions,
desorbing into the gas phase. The product (7) is a chemically which act together (polyfunctional catalysis). Both cases can
and geometrically modified catalyst that does not fulfill the be basically described with more-complex Langmuir–Hin-
condition of remaining unchanged throughout the process. shelwood mechanisms (LHMs), but are seldom considered. A
Neither reaction pathway (B) nor (C) is considered in the simplification restricted to a specific type of active center is
standard model, which leads to the appearance of the “gaps” also common among catalyst manufacturers if they wish to
between model and high-performance experiments. emphasize the specific characteristics of “single-site” cata-
Model conditions are often chosen so that only a minimal lysts.[5a, c, 62d, 137] Their active centers are “all-rounders” that
reaction occurs and the possible readsorption of the products must be active in many different elementary steps.
can be excluded. Despite this, a modification of the cata- We will now formulate a LHM for reaction (18) and
lyst[133] either during or after the reaction was observed by consider only the reactions of the starting materials with each
surface analysis. Remarkably, this change was not taken into other. The intermediate steps of adsorption and activation,
account in the quantitative and theoretical treatment. Either which we discussed in Figure 12, appear in Equations (19) and
the observation was ended after the conversion quantities (20).
were still so small that they could be ignored or they were
regarded as insignificant. This may be motivated by the A þ * Ð A* ; k19
þ 
, k19 ð19Þ
difficulty in determining the exact chemical composition and þ 
B þ * Ð B* ; k20, k20 ð20Þ
the coordinates of the surface atoms in modified catalysts. The
very goal of model experiments is to carry out a catalytic A* þ B* Ð AB* þ * ; k21
þ 
, k21 ð21Þ
reaction under reactions where the coordinates of the
þ 
participating atoms are known or can be determined to gain AB* Ð AB þ * ; k22, k22 ð22Þ
atomically precise insight into the reaction.
The quantitative standard model was derived from the In this notation the rate constants for the forward and
qualitative model shown in Figure 12. This does not need to reverse reactions of the elementary steps leading to the
be completely examined here, as there are textbooks[100, 133a] generation of the products are given. Frequently, desorption
and publications[134] dealing with the subject. Here, the of the products in step (22) can be presumed to be irreversible
intention is to discuss simplifications found in the derivation and thus becomes kinetically irrelevant. This means that the
leading to the quantitative picture. With this we identify, at interaction of the products AB with the catalyst is weaker
least in part, the source of the gaps between the standard than that of the products. This is almost always valid for the
model and experimental results, and we can discuss the CO2 molecule that results from CO oxidation, except if
application of the model to high-performance systems. We carbonates can be formed.
start with the very simple model reaction: More problematic is the further assumption that the active
centers (*) remain unmodified throughout the reaction and
that potentially necessary regeneration steps take place so
A þ B þ cat: ! AB þ cat: ð18Þ
fast that they are of no kinetic relevance. In the case study of
methane activation we saw that this is also applicable, for
The first simplification is that the catalyst, “cat.,” pos- example, in the case of metallic catalysts for CO oxidation.
sesses active centers that react equally with A and B. Thus, However, there is a large and technically relevant class of
there is only one kind of active center that effects both reactions to which this simplification does not apply. These
adsorption and reaction. In the introduction we already are the oxidation reactions. Here an oxygen atom is removed
mentioned the term “structure sensitivity[132d, e, 135]” that stands during the reaction from what is always a multiatom active
in contrast to the “universality” of active centers in Equa- center and/or oxide ions become hydroxy groups by accepting
tion (18). Exceptions to this are reactions whose kinetics protons from the starting material. The regeneration of the
depends on the specific adsorption of a single component (A) active center is seen as rate-determining, while the generation
because the other component (B) adsorbs spontaneously and of the product is not assumed to be kinetically decisive. This
is easily activated. An important example of this is hydrogen is, in any case, what is understood with the term “Mars-van
on noble metal surfaces. Here the generation of a single Krevelen Mechanism (MvK)”. The reverse assumption, that
product depends only on the centers that bind and activate the reoxidation is fast while the product generation is slow,[138]
(A). This case appears in important reactions of organic is also described by this acronym. The apparent contradiction
hydrogenations, the reduction of nitrogen to ammonia, and can be clarified by observing that the reaction rate is strongly
the oxidation of SO2 and CO. Even more kinetically simple controlled by the concentration of water in the reaction
are reactions that are catalyzed by solid acids. They obey mixture and, therefore, depends on the exact conditions of the
either Equation (18) or are, like isomerization, formally even measurement, which leads to difficulties in making general
more simple (A + cat. ! B + cat.). The actual reaction statements[139] about the process. A further discussion can be
sequence and the nature of the active centers are, however, found in Ref. [140]. However, the original publication,[141]
extremely complex[136] and are not discussed further here. from which the process gets its name, does not contain this

3482 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

interpretation. Rather, the authors extended the mechanism precisely if an estimate for the number of static active centers
in Equations (19)–(22) by an additional step that was justified is used, for example, the number of step edges.[2g, 96, 130a]
by the regeneration of the active center. The following set of equations connects the reaction rates
The LHM approach in Equations (19)–(22) also requires for every partial step with the essentially independently
that the stoichiometry with respect to all participating atoms determinable coverage of the relevant species.
and the exchanged charge equivalents remains unchanged.
þ 
This will lead to additional reaction steps in the case of more- r19 ¼ k19 pA q* k19 qA
complex molecules and is the reason why Equations (19)–(22) þ
r20 ¼ k20 
pB q* k20 qB
often describe elementary steps. þ 
ð25Þ
r21 ¼ k21 qA qB k21 qAB q*
Now we will formulate the corresponding reaction rate for þ 
r22 ¼ k22 qAB k22 pAB q*
every step in the LHM approach in such a way that they are
proportional to the total number of active centers. We take
from Langmuirs theory of catalytic reactions[119] that for From this we see that the knowledge about the degree of
every participating species, the number of occupied sites on coverage of the participating reactants, which can be deter-
the surface results from the sorption steady state under mined by surface science experiments makes the equation
reaction conditions. For molecular adsorption of A this would solvable. In this way kinetics can be calculated through
be: experimental results that characterize the sorption of relevant
species by using the many different methods of surface
KA pA science or of other methods in the arsenal of physical
qA ¼ ð23Þ
1 þ KA pA
chemistry and under many possible reaction condi-
tions.[2e, 3c, 22d, 24a, 143] Then we can verify the reaction mechanism
while we obtain
that determines the exact form of the system of Equa-
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi tions (25). After we have identified the reaction mechanism
KB2 pB2
qB ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi ð24Þ with absolute certainty we can, alternatively, assess the
1 þ KB2 pB2 assumption about the nature of the active centers. This is
contained in the values for the degree of coverage obtained
for the dissociation of a diatomic molecule. The quantity q from the specificity of adsorption/desorption. However,
denotes the degree of coverage, and p stands for the pressure whether we can unambiguously obtain a reaction mechanism
or, to be more precise, the chemical potential of the starting from kinetic data has already been called into questioned and
materials and products. The constant K specifies the sorption is, for example, also discussed using the example of the
steady state as the quotient of the rate constants for the oxidation of HCl.[50i]
adsorption and desorption of the material in question. If the sorption characteristics of a reaction system are
In some cases setting the pressure or concentration in measured exactly over a range of reaction conditions, it is
solution can be equal to the chemical potential, which can observed that the degree of surface coverage depends not
lead to significant errors. A drastic example is found in only on the chemical potential but also on changes in the
assuming that ammonia can be viewed as molecular nitrogen specific bonding ability of the catalyst with the surface layer.
when describing a reaction of atomic nitrogen. On many This is closely related to the dynamic of sorption[34b, 144] and
surfaces ammonia decomposes easily into nitrogen atoms necessitates special care during analysis.[145] The result is, also
while molecular nitrogen is much more stable. The effective due to this dependence, that the value of the analytical
pressure for a nitridation reaction from either ammonia or characterization of a reaction mechanism becomes unclear[4b]
from dinitrogen is then different depending on the decom- if a range of reaction conditions (temperature, pressure) is
position constants. Ertl used the term “virtual pressure”[142] to considered and not just one specific combination of these. An
describe this. For the case of Fe4N[142] 3  1011 bar ammonia instructive illustration for this is a study[146] on the synthesis of
has the same effective pressure as 3100 bar dinitrogen at methanol from either CO or CO2 as the carbon source. Under
670 K. conditions of high performance, the carbon source is CO2
Now we can specify the reaction rates for every partial [Eq. (11)], whereas at a 50 K lower temperature and “differ-
step. For this we need the additional simplification that we are ential reaction conditions” CO is the carbon source for
interested only in the number of reactions per unit time and methanol [Eq. (12)]. Different intermediates will be found.
per active center (TOF) and not in the actual number of The controlling factor for this is the surface coverage of water.
centers themselves. This simplification is, however, not The analysis is involved in the context of the present
absolutely necessary because we can formulate the following discussion as several parallel and consecutive reactions
equations in such a way that the total number of active centers (synthesis of methanol, water gas chemistry) contribute to
appears explicitly as a parameter. However, this leads then to the net observed conversion. Furthermore, the assumption
the problems of the absolute determination of the active that all reaction products desorb easily from the surface and
centers described above, which is ignored in many observa- that no readsorptions occur in the reaction system is invalid.
tions involving reaction kinetics. Commonly an approximate Enthalpies of chemisorption as a function of surface
value, the number of surface atoms per close-packed metal coverage can be obtained nowadays[147] with high precision
surface area, is taken for this parameter (ca. 1015 cm2). thanks to single-crystal calorimetry. A precise determination
However, it is possible to formulate this significantly more of energy data can be related to a precise determination of the

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3483
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

surface coverage, including the structure of the adsorbate. gradients appear parallel as well as perpendicular to the
The unfortunately very involved and highly tricky experi- central axis of the reactor and have a significant effect on the
ments are invaluable reference data,[6, 30d, 148] for calibrating chemical potential. Further complications are the inhomoge-
concepts and theory of chemisorption. In some cases such neities on the scale of the granular packing (“split solid”) as
data were used to decipher the reaction mechanism along the well as of the internal pore structure of the material.
procedure outlined with Equation (25). Considerable experimental finesse[155] is needed to exclude
Sorption microcalorimetry on powder samples is notice- all of these factors and obtain kinetic data free from their
ably less precise (in the investigation of the degree of influence. This applies to the synthesis of the samples used
coverage) but still applicable to high-performance catalysts. that must be homogeneous in structure and reactivity as well
The measurement requires extreme caution and experimental as to execution of the experiments themselves. If this is not
precision,[149] but data for kinetic analyses can be obtained for the case and error-laden data are used for the catalytic
a large array[150] of systems. analysis using Equation (25), the resulting parameters will be
As a rule, however, the necessary observations are not largely inaccurate and also, accordingly, the corresponding
available,[2i] but we have only estimated the corresponding conclusions. This situation has resulted in the large “diversity”
parameters. Quantum mechanical methods in combination of “characteristics” described in literature and complicates
with quantitative measurements[151] have been applied with the contribution from dearly won experimental observations
great success in catalysis research to calculate ab initio to expanded molecular knowledge.[71] The examples[13a, 156]
relevant parameters for elementary step reactions. This has show that the required quality of work is indeed obtained
enabled us to establish the solutions to kinetic equations on from a series of reactions in an integrated approach accom-
a new basis.[10c, 152] panied by the corresponding conclusions that are in accord-
Furthermore we know that the sum of all centers must be ance with the principles of chemistry and need no “special
equal to one. To be able to finally solve the system of characteristics” for their interpretation.
Equations (25), we will introduce a series of further simpli- The influence of the gradients throughout the reactor will
fications. We will assume that the reacting system is in be especially drastic if the reaction proceeds very quickly and
a stationary state. In this way all surface coverages and with high reaction enthalpy. Partial oxidations are exam-
reaction rates become independent of time. There is, thus, no ples[157] of such reactions. Data allowing the calculation of the
chemical dynamics from activation and deactivation of the degree of surface coverage and, therefore, input for mecha-
system. This can be ensured in an experimental setting, but nistic considerations are obtainable from measurements of
only with great effort. We also assume that adsorbed species profiles of the temperature and material composition in novel
do not impede one another or tend toward association (see profile reactors[51h] or tap reactors. Such data show how far
Figure 12 (2)). Moreover, we exclude the possibility that the real reactor experiments currently are from the original
active centers differ amongst themselves or when in contact underlying assumptions. Chemical engineering science is
with reactants: the surface is locally homogeneous. engaged extensively in the quantitative treatment of these
Another significant simplification is that we will not make effects; it is important to realize for the current work that it is
any statements about the spatial coordinates of the reaction a considerable task to interpret kinetic data from a catalyst in
rates in a reactor. The reaction at all positions of a reactor a given reaction for the purpose of supplying a basis for
proceeds at the same rate and form the same products a discourse on molecular processes.
because the chemical potential or the sum of the starting Up until now we have assumed that the catalyst itself does
materials and products is assumed to be the same everywhere. not change during the reaction. A well-known example of this
This is where chemical process engineering comes into play. is the catalyst used in ammonia synthesis. However, this is not
Reactors can be planned in such a way that these conditions always the case in the oxidation of CO with oxygen, as was
are nearly fulfilled: such a “stirred-tank reactor” can actually shown spectacularly by Ertl[22d, 158] through the nonlinear
be approximately realized. Much more often, however, in behavior of the reaction at low pressures. Nonlinear changes
laboratory experiments, its idealized form is assumed as also appear at higher pressures in this reaction, which can
a starting point. The model is applicable in situations with even be in part detected as periodic changes of the oxidation
small conversions (differential reaction conditions) so that state of the catalyst.[159] In the meantime, such significant
this simplification can indeed be helpful. This is also the case periodic structural changes[160] have been found in whole
in many model experiments of surface science. If, however, series of reactions.
attempts are made in such experiments to close the pressure Model conditions are often chosen such that only
gap and achieve high reactions rates, significant complications a minimal conversion is obtained and the readsorption of
result with the identification of active surfaces and with the the products can be ruled out. However, this “plausible”
validity of the approximation of homogeneous[153] reaction assumption is often incorrect. It influences the analysis of
rates. sorption and reaction data as can be seen in the example of
These simplifications are, however, not valid in the vast the careful analysis of the sorption of dinitrogen on an
majority of all practical reactors and laboratory experiments ammonia catalyst (which would be referred to as a chemically
aiming at measuring catalytic performances.[13b, 154] Rather, “harmless” case).[4a, 141, 161]
considerable gradients in the chemical potential are present Despite being under “model conditions” a change in the
along the typical packed bed of a catalytic experiment. As catalyst during or after the reaction was found by surface
many reactions have an associated heat exchange, energy flux analysis. A concept central to this is the adsorbate-induced

3484 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

restructuring.[133b,c, 162] By using surface-sensitive methods Table 3: Some parameters[164a, 164f ] for the sorption of ethylbenzene (EB)
before and during or after the reaction, it could be established and styrene (St) on single-crystal model catalysts.[a]
that the surface structure of the catalyst was significantly Surface Edes St Edes EB q EB q St q St/
altered. This has considerable consequences for the adsorp- [%] [%] q EB
tion ability (sticking coefficient, geometry of the adsorption Fe3O4 (111) 118 86 92 100 250
centers) and, thus, for the catalytic reaction itself. This Fe2O3 (111) 73 64 37 29 0.8
observation is a deep-seated concept in the surface science KFeO2 (111) 65 65 25 6 0.2
of catalysis and explains, as given above, the necessity of [a] The desorption energy is given in kJ mol1. The degree of surface
in situ structural investigations. The simplification given ear- coverage was obtained with a reactant pressure of 100 mbar at a reaction
lier that the active and initial states of a catalyst are the same, temperature of 900 K.
the latter being well understood, is not applicable in many
cases.
According to the procedures discussed above it would be
important at this point to measure the kinetics of the reaction
9. Case Study of the Dehydrogenation of Ethyl- of a model system experimentally under relevant conditions
benzene over Iron Oxide (atmospheric pressure, 900 K, excess of water vapor). How-
ever, this turned out to be a complex challenge that could only
The dehydrogenation of ethylbenzene (EB) to styrene be solved after significant methodological development-
(St) is a large-scale technical reaction for the manufacture of s.[166a,h, 168] The most important results are given in Figure 14.
a monomer for polymer synthesis. It takes place using iron The first step was to build a microreactor[169] from materials
oxides as promoters at high temperatures (873 K) with the that could withstand the conditions of the reaction and enable
addition of a 10-fold excess of water vapor. This is an the transfer of the model sample into the reaction atmos-
endothermic reaction (DHf298 =+ 123 kJ mol1). phere. To eliminate transport limitations, a reactor concept
was chosen that was able to reach the required temperature
by laser heating. The entire apparatus then needed to be
C8 H9 ! C8 H7 þ H2 ð26Þ
integrated into a ultrahigh vacuum (UHV) set-up to be able
to perform the necessary structural analysis and manufacture
The analysis of the role of the catalyst[164] shows that the the samples with the required quality. Figure 14 A shows
active phase is a ternary iron potassium oxide (KFeO2) that is schematically the construction of the microreactor.
metastable under the reaction conditions and becomes Fe3O4 It was seen very quickly[166g] that the single-crystalline
and KOH through a complex reaction sequence that may be surface showed no reactivity in the target reaction. We
influenced by promoters. Parallel to this, the reaction suffers observed[166d] an induction period after which the activity grew
from carbon deposits resulting from the polymerization of the strongly. The analysis of the samples after the reaction
products.[165] This reaction is limited by the on-line gasification showed very clearly that the single-crystalline structure was
of carbon with water vapor, which is supported by the lost and that small and rough oxide islands had formed from
potassium promoter. the initial thin and flat layers on metal substrates (Pt, Ru) that
To decode the complex relationship between the dynam- became covered with carbon during the experiment. LEED
ics of catalysis chemistry, the chemistry of unwanted deposits images in Figure 14 B give an impression[166a] of the massive
and the actual target reaction, an extensive campaign[166] was reconstruction: after the reaction only the weak reflections of
mounted for the manufacture and application of planar model the metal substrate can be seen.
catalysts. Another goal was to answer the question of why an The Fe3O4 phase turned out to be completely inactive,
“organic” reaction must be carried out under such harsh although in real samples it is the predominant phase after the
conditions. Thin single-crystalline films of model oxides were reaction; it is a deactivation product. The Fe2O3 phase is, on
made and characterized structurally.[167] The experiments the other hand, very active, as can be seen in Figure 14 C. The
were used to determine the adsorption parameters of starting initial rates that were observed for the target reaction[170] are
material (EB) and product (St) with these samples without approximately 1000 times higher than is measured on tech-
problems related to transport limitations. These data enabled nical samples under stationary conditions. The model reaction
an estimate of the actual coverage under reaction conditions, shows us in a somewhat unexpected way that technical
and the results are given in Table 3. catalysts have still significant potential effectivity in typical
The data show very clearly why the ternary oxide is tubular reactors even without a phase change away from iron
superior: the potassium is not only a promoter but also an oxide. The reason is probably found in the instability of the
essential co-catalyst that mainly ensures desorption of the active phase at the local chemical potential (too reducing) in
product and, thus, access to the active sites in addition to the tubular reactor. The measurement in Figure 14 C1 shows
serving many other functions. At the same time a clear and that the phase is not stabile and that it develops into a stable
desirable excess of the starting material relative to the permanent state with low activity; in agreement with the
product exists on the KFeO2 surface that minimizes not results from other groups[171] this was identified as an oxide
only unwanted polymerization but also improves the reaction phase completely covered by carbon. It can be concluded
rate because the likelihood of contact between the starting from this that the carbon deposits lead to a modification of
material and active center increases. the catalyst that, after a short highly active period in the 10-

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3485
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

An extensive amount of new chemistry[37c, 172] results from

this picture. It leads to a new concept of active centers for the
oxidative hydrogenation of EB to St on nanostructured[173]
carbon. By minimizing the transport barriers for material and
energy (nonporous catalysts,[171] heat conduction of graphite)
a stable performance[172a] and a high selectivity of the reaction
could be obtained, even in the presence of large amounts of
oxygen. In this reaction, process water does not have to be
added and the reaction temperature can be reduced by 300 K.
If a small amount of oxygen is added to the reaction
mixture in the dehydrogenation reaction over iron oxide
(1:10), then a very stable and high activity[174] is observed
(Figure 14 C2) that was found to be reversible by turning the
oxygen source on and off. The scale of the addition of oxygen
was chosen according to Equation (26), so that hydrogen from
the reaction could be burned and the reduction of Fe2O3 to
Fe3O4 was minimized. In this way, the generation of carbon
was completely eliminated, with the reasons, not com-
pletely[175] understood, being found in the defect structure of
the oxide. From Figure 14 D it can be seen that this conclusion
from a model system can be reproduced[174] quantitatively on
a real-world polycrystalline system (a pressed powder pellet).
Furthermore, it was shown[170b] that the addition of
potassium to iron oxide has the same stabilizing effect as
the addition of oxygen: the catalyst is significantly more
stable against reduction. However, the potassium phase is not
stable against hydrolysis and, therefore, loses its effectivity
with time, as has often been seen in real-world systems.[164b,c]
The beneficial effect of adding steam to moderate the
reducing effect of the product water has a simultaneous
detrimental effect on the stability of the bulk catalyst phase.
Such antagonistic behavior is typical in complex empirical
catalyst formulations without a clear understanding of the
functional interrelation of the components. The model inves-
Figure 14. The dehydrogenation of EB to ST. A) Microreactor for the tigations show that the effect of the potassium can be imitated
investigation of single crystals. B) LEED images (75 eV) from a Fe2O3 by the addition of the “co-catalyst” oxygen that eliminates the
(111) surface before and after a reaction at 873 K with EB. C) Changes
damaging hydrogen to avoid a destabilization of the main
in the rate of EB consumption over time on a single-crystalline Fe2O3
catalyst; (1) only EB with water (1:10), (2) with EB and oxygen catalyst. The implementation of these model results in a real-
(1:10:0.5). D) As in (C) but with a polycrystalline Fe2O3 catalyst world situation that is demanding because transport processes
(optical image). in actual large reactors, discussed above, as well as oxygen
dosage in amounts ensuring that organic species are not
fold excess of water, is composed only of carbon. The burned are problems that are not solvable at this time. A
instability of the reactor operation that is often observed in completely new development of the reaction process would
practical operation can be clarified as an on-and-off behavior have to be started and the catalyst would certainly have to be
of the catalyst between an oxidic highly active state and equipped with oxygen storage properties to make it stable
a carbon-containing state of low activity. against local fluctuations in oxygen partial pressure. An
The product, hydrogen, controls the redox state of the alternative would be the realization of carbon-based cata-
catalyst with the admixture of water working in opposition. lysts[37b, 176] whose specific activity[173] need, however, to be
This occurs through a reduction of the chemical potential of improved.
hydrogen and because the water acts as a source of oxygen The depth of the qualitative and quantitative insight
through catalytic splitting of water. This oxygen, present in already gained was used to design a microkinetic model[177] of
small quantities, does not burn the product St, but instead the reaction based on experimental observations. In contrast
gasifies the deposited carbon. The high reaction temperature to earlier attempts that followed the approach given above
can be explained by two factors. First, the activation of the with an active center of unspecified nature, in this study
solid-state reaction for the in situ creation of the metastable a model of a catalyst is used that boasts both oxidic as well as
iron–potassium phase and second the necessary high reaction carbon-based active centers with the generation of these
temperature for water splitting and the gasification of carbon. active centers explicitly taken into account. A verification of
The endothermic character of the target reaction does not the conceptual veracity of this model was obtained by the
justify the high temperature. successful description of the forced change in the catalyst

3486 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

from oxidic to carbon-based through a nonstationary reaction development of catalysts with high performance. Such
process. systems correspond weakly to the boundary conditions of
The example shows that valuable, fundamental insight model catalysis and their theoretical analysis in the frame-
into complicated technical reactions can also be obtained work of the LHM approach. Therefore, with few exceptions,
through model experiments. Apart from new knowledge, this such as ammonia synthesis,[2j] essentially no general relevance
insight also provides a rational basis for new fundamental of the results from model experiments can be expected. Two
developments. The latter may be motivated by the realization strategies have been developed to close this gap. One
that the current catalysts are lagging far behind the ideal approach uses traditional surface science and extends the
possible performance and lifetimes. However, significant analytical methods to application under high pressures from
efforts in synthesis as well as in functional verification of several mbar to several hundreds of mbar. For this purpose,
the realization of the concept for a given catalyst will be typical methods such as atomic force microscopy, sum
required. frequency microscopy, and NAP XP spectroscopy are
employed.[24a] Model catalysts in the form of single crystals
were used and experiments were developed for their inves-
10. Catalysts Are not Static tigation under relatively high “ambient pressures”
(200 mbar). Significant structural changes[163a, 184] of the
If we return to our fundamental considerations, we see model systems were observed with differing interpretations
that the understanding of dynamic proceses in catalysts has given.[185] In a well-known case with CO oxidation at high
been analytically ensured,[21a, 178] but for the most part have pressures[163, 184c] over Pt crystals it could be shown that the
not been treated in a general quantitative and theoretical way. changes were due to the introduction of small amounts of
The prominent exception, which is widely regarded as such, is impurities, such as water, through the reaction gases and not
the treatment of the dynamics[21i, 158b] of the CO oxidation because of the pressure of the reactants per se.[153a] However,
under the special conditions of a periodically oscillating the observation of the effect of the trace gases was important
reaction process. In the general catalysis science no con- because it was understood that the stability of the reacting
sequences were drawn from this; either the observations were surface is determined by many factors that are often not easy
stopped after such small amounts of conversion that the effect to control. A further uncertainty develops if structural
of the change was ignored or the dynamics were seen as analyses under ideal very clean conditions are to be trans-
insignificant. On the other hand, a prominent excep- ferred to less-well-defined environments, for example, during
tion[11a, 21a,f,g,j, 179] to this assertion are the catalysts for partial a reaction with high conversion. NAP-XP has developed from
oxidation that change profoundly[140c] under the conditions of a curiosity[186] to a well-established method[187] that is found at
their use and can reversibly offer oxygen from their bulk or many radiation sources around the world as well as in
surface.[180] laboratory experiments.[188]
One of the reasons that dynamic effects are so under- The other approach is to match the reactivity of the
represented in kinetic treatments of heterogeneous catalysis model system with that of a realistic high-performance
is that the theoretical and quantitative treatment of such catalyst. For this, formidable challenges must be overcome
changes requires an exact description of the atomic structure to deposit nanostructures of the relevant active components
of the modified catalyst. This boundary condition has a greater in a precise way on well-defined model surfaces of typical
limiting effect than the technical difficulties on the transfer of catalyst carriers. These include the oxides of Si and Al that are
know-how from model conditions to high-performance con- not easy to study with surface-sensitive methods because of
ditions in atomically precise catalysis research. However, if their electrically insulating properties. The active components
this transfer is actually successful, precise analyses grounded may be either metals or the oxides themselves. This approach
in simple facts can be obtained for seemingly complex was very fruitful for understanding the surface physics and
observations.[6, 14, 30d, 83a,b, 144, 181] It is an invaluable advantage of reactivity of “complex” systems such as Pd/Al2O3,[6, 181c, 189]
these studies to be able to greatly simplify the phenomeno- Au on diverse carriers[14c, 25a, 31b, 190] or supported vanadium
logical diversity of possible explanations, such as “special oxides.[161c, 191] Catalytic reactions were observed, such as
surface state” or “remote control of the reactivity”.[182] the oxidation of methanol or the hydrogenation of alkynes,
Neglecting structural changes is usually accepted as an under conditions that permitted an analysis of the structure
approximation, as is limiting the investigations of the of the active components. This was possible because the
reactivity to such an extent that at least no change of the nanostructuring enabled an increase in the reactivity
bulk structure occurs. This is not to be confused with the very relative to macroscopic crystals of the same substance. An
deliberate adsorption-induced reconstruction[133a] of metal important step is the realization of model systems as thin
catalysts. Conducting model catalytic studies under the most carriers of an oxide film on an underlying single-crystalline
realistic conditions possible while stopping the chemical metal carrier with nanostructures as the active com-
dynamics[183] is very well justified by the gain in quantitative ponent having a very narrow distribution of morphological
insight. However, the results of these studies should not be characteristics.[27a, 181b] It allows the application of scanning
applied lightly or uncritically to high-conversion catalysis or probe methods for direct imaging and simultaneously
high-performance catalysts. the use of vibrational and electron spectroscopy for the
Thus, a gap appears in the transfer of knowledge from analysis of the chemical structure. The experimental
model experiments to the analysis and also the further and theoretical application of the knowledge gained

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3487
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

from this approach is currently[6, 30d] being extended even This assumption[187a] is, however, acceptable for limiting cases
further. with very small conversions and simple reactions that lead to
thermodynamically stable products. If the local chemical
potential at the site of the active center is increased to the
11. A Critical Survey: Dynamics in Catalysts point that conversion rates rise significantly, new reaction
possibilities between reactants and the active center appear,
The meaning of the standard model for the development as indicated in Figure 12 B,C . The catalyst will then change its
of the understanding of the steps making up heterogeneous surface during conversion, in which case an appropriate
catalytic processes[22d, 24a, 192] cannot be underestimated. The regeneration step with corresponding elementary steps must
core message is that the complete equation of a catalytic be included. More problematic for the standard model is the
reaction can be described by a series of intermediate steps case that the catalyst becomes profoundly altered and a new
consisting of adsorption, reaction, and desorption. Every active phase results from the starting phase originally found in
intermediate step is characterized by a number of elementary the reactor. This new phase may only be stable under the
processes with rate constants for the forward and reverse given chemical potential of the reaction and have its own
reaction as well as surface coverage. The specificity of the unique surface structure. Under different potential conditions
chemistry between the catalyst and the reactant is expressed (such as at room temperature in air) it may be metastable and
by the degree of surface coverage. This is typical for the cause the catalyst to decompose or it may bring it back into its
surface termination of a catalyst and is understood as original form. Such significant changes appear either for the
a function of the reaction temperature. A change in the structure of the active phase[194] or they can also occur in
catalyst caused by the reaction beyond the state of being relation to the entire chemical composition.[45b, 51j, 61, 195] We
“covered” or “not covered” with differing surface structur- differentiate the cases in which the bonds are formed between
es[21i, 158b] is excluded. Therefore, no reaction steps are needed reactants and catalyst precursor (for example, hydrides during
for regeneration of the active centers after the catalytic the reaction with metals[196]) or where the catalyst decom-
reaction. This theory can, therefore, center on the processes of poses and a volatile component is either added[51f, 197] or
the desired transformation of the starting materials through permanently removed[195a, 198] (this applies for oxides and
transition-state theory as the link between chemistry and nitrides). Often complex oxides decompose during this
molecular elementary processes. The exact chemistry of the process into thermodynamically stable binary oxides that
catalyst can be disregarded here because the numerical values irreversibly destroys a catalyst. Such processes are very
for sorption and reaction rate constants are encoded with this prominent[199] in contaminated, complex oxide phases[200] in
information. partial oxidation while pure-phase materials are significantly
If calculations are performed for a relevant reaction it is more stable[201] against these types of complications.
quickly seen that a substantial complexity results from the Figure 15 a summarizes a number of influences that affect
number of necessary elementary steps. The Equations (25) the degree of surface coverage—the central quantity for
illustrate a very simplified model situation because in practice catalysis chemistry in the standard model. It follows from the
nowhere near all of the parameters that are needed can be standard model that the chemical potential and the sticking
determined. Instead, they are estimated or critical values are coefficient are of fundamental importance. However, both
obtained theoretically; experimental measurements can only are subject to a series of influences that are considered to
be used in rare cases. The consequence is that there are many remain constant or are even ignored as a simplification in the
combinations of equations and estimates of parameters that standard model. In addition to the macroscopic variables
describe an experimentally observed performance of a com- pressure, the temperature and composition of the reaction
plete reaction equation. Thus, a kinetic model of a reaction mixture (the latter of which does not change in the standard
can make a reaction mechanism plausible but not render it model when viewed as a limiting case with small reaction
a certainty. This does not even apply to the reaction network, conversions), the chemical potential is also determined by the
that is, for the sequence of complete reaction equations that reaction conversion and the material and energy transport
are possible to arrive at the products from a combination of characteristics of the system. The chemical potential is then
starting materials. The example of the activation of dioxygen dependent on the position, a “local” chemical potential, and
as a very simple reaction, shown in Figure 2, illustrates this the kinetics can no longer be taken to be independent of the
well. The seemingly simple example of dimerization of exact reaction position (“mean-field approximation”).
methane to ethene and hydrogen (see Figure 1 and the The phenomena of catalyst dynamics, which are neglected
MgO case study) is a dramatic example of the complexity[2b] by the standard model, influence the sticking coefficient and
of the hidden, underlying reaction network. It is still not the local chemical potential in an integral way. The influences
completely clarified to this day. Furthermore, the example of “surface structure” and “molecular dynamics” given in
the generation of methanol from CO2 and hydrogen conceals Figure 15 can be corrected by an adaption of the standard
the extensive complexity in a theoretically calculated model to suitable values of the chemical potential. The effects
model[193] of the catalyst Cu/ZrO2 that treats all the possible represented by the yellow boxes in Figure 15 contradict the
reaction pathways. simplifications in the standard model and the LHM approach
Here it is of course understandable to not further and are, therefore, not accounted for. These influences cause
complicate the situation by relinquishing the concept of a feedback loop between the structure and catalytic activity
a rigid active center that may only be occupied or unoccupied. on the one hand and the local potential on the other. With this

3488 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

possession of the many material–chemical observations

needed to be able to a make an assessment of this chemistry.
This situation is in sharp contrast to the insights into reaction
pathways of the starting materials and products that we are
able to evaluate well with molecular chemistry.
The task is made even more difficult because the local
chemical potential is the decisive factor and not the potential
of the reactants at the entrance to the reactor. The chemical
potential is subject to gradients in a reactor due to the
progress of the reaction (macroscopic), transport on the scale
of boundary layers of material flow, particles and pores
(mesoscopic), and the distribution of active centers in the
active phase (microscopic). A similar hierarchy results for the
transport of the reaction energy. Both hierarchies, given in
Figure 1, determine the local potential.
Unfortunately, this is still not complex enough because the
response of a solid body to the local chemical potential is
determined through solid-state kinetics. This is dependent on
the dimensions of the reacting particle (size, form, material
transport mechanism) and is an often ignored motivation for
nanochemistry[27c, 37p, 48, 202] in heterogeneous catalysis. The
defect chemistry that we sometimes control with promoters,
but usually regulate[203] unconsciously with synthesis proce-
dures, explains the molecular dimension of the response of
a catalyst to its environment. In semiconductor[204] catalysts,
Figure 15. Influential factors that control the degree of surface cover- additional size-dependent changes of the electronic structure
age of a material in a heterogeneous reaction in the limiting case of
caused by boundary layers come into play. Their range is often
high-performance conversion.
comparable to the size of the particle and, therefore, leads to
significant changes in the electronic structure of a working
they contradict the single-crystal approximation that the catalyst as compared to a macroscopic sample. In this way the
catalyst can be investigated essentially independently of its morphology, carrier–metal interaction, and the actual struc-
reaction environment and in the form of a model in such ture of the active phase determine the response of the catalyst
a way that relevant parameters for the elementary steps of to the local potential under reaction conditions. Every
a kinetic description could be obtained. Unfortunately, this is deviation from the assumed temporal stability of the fluxes
not applicable in the general case as will now be discussed, of material and energy complicate the problem even more.
and this leads to the insufficient assertiveness of a conven- The influence of the complexity of the controlling factors
tionally determined, self-explanatory prediction of catalytic on the events in a reactor is shown in Figure 16. Central to
effectivity. This deficiency in no way means that there is no Figure 16 is the cycle of processes (blue) that describes the
general physically based concept of heterogeneous catalysis, formation and effectivity of the active centers. The macro-
but rather that the standard model is simply not yet scopic variables are colored green and are treated in the
constructed in a general enough way. standard model and by chemical engineering. They determine
After the above discussion, the description of a heteroge- the macroscopic effect of a “black box” called a “reactor”, in
neous reaction in its reaction network must also incorporate which a targeted conversion takes place. Material chemistry,
the reaction of the catalyst with the reactants beyond which transforms the catalyst precursor into active centers
adsorption and activation. The complexity of the task then and also is responsible for their regeneration (the entire “blue
increases and it would seem to be necessary to accept the cycle” in Figure 16), is not treated in classical physical
aforementioned simplifications. However, through this the catalysis research. Rather the focus is on the chemistry of
option is given away of making conclusions about the material the conversion of the reactants and leads to the “gaps” in this
chemistry of the catalyst from observing its performance. The scientific field. The events become complex because the
assumptions about the material character of the active centers different steps of this reaction cycle of the catalyst can be
used in the standard model and in derived empirical concepts controlled by different influential elements.
of catalytic reactions are really only speculative and are not These factors are themselves not static parameters but
based on observations. The case studies of OCM and of EB rather dynamic processes (yellow in Figure 16). This means
dehydrogenations may serve as examples. This is a central that they must be treated as fluctuating with time. A good
weakness because the model of the reaction says nothing definition of an influential factor in material chemistry is,
about the dynamics of the catalyst in the chemical potential of therefore, a quantity that fluctuates around its own static
the reactants. In general we do not know the chemical mean value. These dynamics are coupled with one another
dynamics of the catalyst either at the level of the active center because a superposition of external quantities and the
or of the active phase. Regrettably, we are not even in productivity of the catalyst control them simultaneously.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3489
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

“ligand-exchange process” and “tautomerism”[206] are the

corresponding terms that are well established in chemistry. In
this case it is new that these processes, which are defined in
molecular chemistry through the reaction system, are also
controlled by material chemistry and conversion in a hetero-
geneous reaction involving a solid that was hitherto consid-
ered as “nonreactive”. An example of this dynamic could be,
according to Equation (27), the fluctuations of a metal–
oxygen double bond into a radical structure with single bonds
and a reduced metal center.

Mnþ ¼O Ð Mðnþ1Þþ -O* ð27Þ

Such a process illustrates in a straightforward way how

adaptive metal centers, acting as redox centers, activate
a nonreactive substrate without further activating the prod-
uct. The formation of an oxyl radical[207] is discussed in the
molecular chemistry of oxygen activation and was identified
in molecular complexes.
The morphological dynamics that determine the real
structure of the catalyst during synthesis and under reaction
conditions is well-known from solid-state chemistry and
studies on the generation of solid bodies from fluid precur-
sors.[39c, 208] The basis here is the influence of nanochemistry on
catalysis.[48] These dynamics describing the assembly of a solid
body as a fluctuation about a translationally symmetrical ideal
structure is mainly controlled by the synthesis recipe[193b, 209]
for the catalyst precursor. Often considered the scientific
embodiment of the “black magic” of catalyst synthesis, it
actually represents a sequence of completely understanda-
Figure 16. A realistic description of the feedback loop between the ble[109a, 195a, 209f, 210] kinetically controlled chemical processes.
material chemistry of a heterogeneous catalyst and its function. The
Controlling the morphology and defect structure through
coloring elucidates different levels of chemistry and is clarified in the
text. a variation of synthesis parameters represents yet another
wide and only somewhat understood[5c, 118c, 211] field of catalyst
research.
One of the least well-known areas in catalytic action
(Figure 16) is that of chemical dynamics. In this dimension,
Molecular dynamics[2e, 11b, 130a, 178a] are the most well-under- the chemical composition and the surface structure of
stood part of physical catalysis research. They describe the a catalyst fluctuate together about the mean value of a basis
generation and interaction of adsorbed and activated reac- structure without actually taking on this exact value. This
tants on the sample surface as well as the dynamics of the would then be the thermodynamically stable structure (as is
molecule in the reactant phase.[13b] The field of nonlinear often assumed in theory because the atomic coordinates are
dynamics of catalytic reactions[21i, 205] also belongs to this then fixed). Processes of segregation or deposition, melting of
discussion as it gives us an impression of the power of the the surface, and the incorporation of atoms under the surface
feedback loop to modify the events of the reaction beyond the or into the bulk of the bulk phase can all cause these
picture derived by the standard model. Here it is new that the dynamics. The generation of near-surface substoichiometric
dynamics of the reactants during adsorption are not deter- bonds is normally[212] not observed in a vacuum and can often
mined by the structure of the catalyst precursor but rather not be identified with surface spectroscopic methods. There-
they change with the evolution of their chemical reactions fore, this has often been erroneously referred to as the “dirty
between themselves and with the catalyst precursor and side effect” of bad experiments (that do indeed exist).
become, therefore, a function of the conversion and of the Nevertheless, it is one of the central mechanisms for
quality[2e, 24b] of the catalyst precursor. Typical examples are generating active phases.
the manifold effects that temperature of the reactants has on Often the active phases decompose when the chemical
their reactivity and the dependence of the reaction rate on the potential comes close to the standard potential and leaves
degree of surface coverage (that is actually not included in the behind a modified structure of the catalyst precursor: they are
LHM approach). practically “invisible” in analytical methods that are not
The concept of bonding dynamics that changes an active carried out under reaction conditions. A series of observa-
center into a reactive center through fluctuations is very well tions of such chemically dynamical systems is reported in
known from molecular chemistry and homogeneous catalysis: Table 4 from our studies.

3490 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

Table 4: Examples of chemical dynamics in heterogeneous catalysts. The remains constant. We surmise that changes in the sequence of
references cite sources in the literature that provide context. the steps[228b] of the reaction occur because the catalyst
Reaction Basisphase Active phase Reference modifies its surface chemistry along with the chemical
potential of the reactants.
ammonia synthesis Fe Fe18N1x [213]
formaldehyde synthesis Ag AgsubO [214] The four governing factors represented in the corners of
formaldehyde synthesis Cu CusubO [215] Figure 16 (purple) control the local geometric and electronic
methanol synthesis Cu CusubO + ZnOgr [56e] behavior at the site of the active center as well as the influence
ethylene epoxidation Ag AgsubO + O [216] of material and energy transport. In the standard model these
ethylene epoxidation AgCux AgsubO + CuO1x [217] parameters are seen as constant during the reaction at every
selective hydrogenation Pd PdsubC [218]
observed position of the catalyst and are, therefore, not
selective hydrogenation PdGa Pd@PdGa [56d]
selective hydrogenation Pt C@Pt [219]
explicitly treated kinetically. The consequences of the recip-
formaldehyde synthesis RuO2 RusubO [220] rocal feedback loop and their effects on the dynamics
CO oxidation Ru, RuO2 RusubO [221] described in the standard model are, thus, suppressed, thereby
styrene synthesis DH Fe3O4 KFeO2 [164a] resulting in the “material gap” and the “complexity gap” as
styrene synthesis DH Fe2O3 C@Fe3O4 [170b] well as a part of the “pressure gap”. A path then opens up for
styrene synthesis ODH C CxHyOz [37e] improving the standard model, at least in the choice of the
butane to MSA VOP2O7 VxOy + H3PO4 [222]
approach, to close these “gaps”. This path requires enlarging
butane to MSA V2O5  H2O VxOy [223]
propane to acrylic acid MoNbVTeOx VxOy + TeO2 [224] the scope of elementary steps needed in a microkinetic model
propane to CO NiO NisubO [160b] by specific reactions between the catalyst and the reactants.
Some of them may be considered as being in equilibrium
during steady-state operation, but some of them will be
irreversible as they describe the maturation of the precatalyst
An example of the essential function of chemical dynam- into its active form through the reaction of reactants with the
ics is the effect of Pd as a hydrogenation catalyst in organic precursor.
synthesis. The catalytic effect is not only due to the metal but We recognize that the description of a working catalyst is
also the hydride-subsurface phase,[189a,b] whose exact stoichi- not only made difficult by the structural complexity of
ometry[225] determines the reactivity. Without the generation a catalyst precursor in comparison to a single crystal of the
of the hydride phase, Pd[181c] is not effective as a hydrogenation same phase, the latter of which is used to justify the choice of
catalyst. macroscopic single-crystal surfaces[97a, 127] in the development
It is notable that the relevant catalysts can also be of the standard model. In addition, there is an entire series of
obtained by impeding chemical dynamics. Examples of this dynamics on different scales of space and time that are also
are found in the class of intermetallic compounds[226] that influential. This understanding is not new in principle, it was
prevent the inclusion of reactants[218] in their bulk by means of already formulated by Ertl[97a] in a review article. What is new
stable lattices and, therefore, hinder the action of correspond- is the insight into how drastic the interplay of dynamics
ing reaction-induced modifications on their geometric and changes a catalyst depending on its reaction environment.
electronic structure. For this reason this class of materials is The “exact investigation of active centers” required by Ertl in
predestined for the development of catalysts supported by his article is pushed to its limits experimentally and concep-
theory, which up until now has not been able to incorporate tually if states of the catalyst are considered that deliver
chemical dynamics. significant performance in reaction networks. The surfaces
A limiting case where the chemical dynamics are are then “flexible,” as formulated by Somorjai.[133b] Model
restricted rigorously to the surface, is the possible strong systems[111b, 192] fail in their role as samples with known atomic
bonding of a reaction product to the catalyst, which leads to coordinates because they either cannot reach such levels of
self-poisoning. This case of “autoinhibition” is not rare and performance or because they change into undesired, complex
can result in complex kinetic phenomena such as the well- systems during the reaction, for example styrene synthesis
known rate oscillations[21i] and “compensation effects”.[227] through dehydrogenation. If the simplifications from the
It is also shown in Figure 16 how the dynamic phenomena standard model for the analysis of the phenomenon “hetero-
are connected to each other by control parameters in geneous catalysis” (the “Langmuir period”[111b]) and for the
a feedback loop. This makes it necessary to study catalytic material description of a catalyst[38, 92b, 229] had not been
phenomena under reaction conditions as well as with a wider applied, the research in heterogeneous catalysis would still
scope of reaction parameters such as structural parameters of be stuck in the era of black magic and trial and error.
the catalyst (defect structures, morphology, nanostructures, We return now to the types of dynamics that determine
size distribution, as examples). This is needed because it may the function of a catalyst. They act on different scales of space
be assumed that different controlling factors affect the and time and, according to Figure 17, complete the picture of
observed catalytic behavior in different ways under divergent the “multiscale problem of catalysis” introduced in Figure 1.
starting conditions. An instructive example for this is the Also shown in Figure 17 are the charge-carrier dynamics,
variation of the activation barrier for the oxidation of propane a universal characteristic of solid-state matter arising from its
to propene over a mixed oxide catalyst.[228] This “apparent” basic electronic structure. In the form of semiconductor
constant changes its value by a factor of two if the ratio of properties from many active phases, these have an important
water to oxygen is changed while the amount of propane and not yet fully understood effect[80c, 204b] in catalysis because

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3491
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

then results from the precursor under reaction conditions with

participation from the reactants. This finds its parallel in
molecular chemistry in a solution of central ions and ligands.
In the heterogeneous case, the precursor has the task of
making the components of the active phase available and
stabilizing their nanostructures to such an extent that, with
low activation energy, they display dynamic characteristics
leading in turn to active centers. Simultaneously, the pre-
cursor, which has now become a matrix, ensures that this
dynamic does not turn toward a decomposition of the active
phase into a static, thermodynamically stable state that is no
longer catalytically active.
One deficit in catalyst research is that we simply do not
know enough about these reactions and the mechanisms that
control them. This is due to the nature of active phases being
thin layers or nanostructures, which are difficult to analyze
under normal experimental conditions. In addition, few
Figure 17. Different dynamic control elements affect heterogeneous researchers are interested in the reactivity of nanostructures
catalysts in different dimensions of space and time. See also Figure 1. under the conditions of a catalytic process. One group of
special relevance here, the substoichiometric compounds, has
only been studied in detail with respect to structural
they directly influence the degree of coverage of the reacting characteristics, while studies of the dynamics of their for-
surface. This effect cannot be considered to be independent mation and reactivity are still rare. Active phases also develop
because it is influenced in many ways by the other factors of through segregation and reconstruction, often induced by
catalyst function; it acts rather in a combined[51i, 204a] fashion chemisorption or the dissolution of reactants in the catalyst.
with the other elements of dynamics. In metals the situation is These kinds of processes play an important role in other fields
somewhat easier as the bath of free electrons at the Fermi of material science where they are also well-known[124a, 208d, 237]
level can be assumed to be a reservoir for redox equivalents but the transfer to the structures and reaction conditions of
required for surface chemical reactions. Thus, the electronic catalysts is still missing. An example is the defect chemistry of
structure of a metallic solid is a critically required, and for oxidic materials that is generally very well studied. In
many surface processes, sufficient descriptor. This forms one electrocatalysts and in oxidic mixed catalysts[238] this defect
basic ingredient into the success of theory in explaining and chemistry is thought to also play a decisive role, but is very
even to a certain extent predicting catalytic reactions (see hard to observe under reaction conditions. Therefore, despite
below). insistent calls[239] for increased cooperation between these two
Figure 17 illustrates clearly the large challenges in fields, little progress has been made on this front.
describing the factors affecting a catalyst and, therefore, Active centers are generated from the active phase by
also its design. Many different dimensions in space and time dynamic processes. These are not continuously present in
require analysis that leads to the demand for diverse general but rather result from fluctuations in the material
analytical methods seldom found in close tandem. Figure 17 structure and in the chemical bonds of the active phase. They
demonstrates, therefore, the necessity of cooperation in are rare events on the time scale of molecular dynamics (fs–
catalysis research if we desire to reach the level of under- ps), which clarifies the fact why they exist as high-energy
standing required to make predictions about the functionality states. We may presume that the observed reaction temper-
and stability of a system at high performance. This applies to ature is responsible in many cases for making such uncommon
the required combination of synthetic and functional experi- events possible. It is further quite conceivable that the high
ments, the modeling of kinetics[2a] and to the theory of mean kinetic energy of the system is required both for the
chemical reactions.[11a, 13b, 152, 235] generation of active centers and for regeneration of the
centers after a successful target reaction. This leads to the
energetic control of the dynamics of adsorbate complexes.
12. The Unified Concept of Catalysis It may be speculated whether other forms of energy
supply to a catalytic system, such as photoactivation,[240] the
Let us assume that we have convinced ourselves that the generation of electric fields in electrocatalysis[159b, 241] or the
limits of Langmuirs concept and of the standard model do stimulation from radicals in plasma catalysis,[242] can also lead
not allow us to describe the most essential characteristics of to such active centers directly or with a low additional energy
the dynamic control of catalyst functionality. Then we can ask input. Much more often the effect on the molecular structure
anew the question about the common definition of catalytic of the reactants[243] by nonthermal energy, induced directly or
function common to both the molecular and interface- through the catalyst, is investigated as a central catalyst
determined systems. The synthesis of heterogeneous catalysts, function. The discussion of “confinement effects”,[244] now
according to Figure 16, does not supply the active functional in vogue again, is an example of this, although here a static
material itself, but rather only the precursor. The active phase catalyst is assumed. The resulting complexity of a connection

3492 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

of such effects with the structural dynamics of active phases

will not be discussed here.
The concept that active centers are in general high-energy
sites in an active surface phase created during catalyst
synthesis is not correct. Rather, the synthesis enables the
reaction of the catalyst with the reactants through defects and
nanostructures. The result is the creation of the active phase
and finally the dynamics of the formation of the active sites. In
heterogeneous catalysts that have active centers that are
difficult to define geometrically[232, 245] (how many atoms are
participating at any one time?) the combination is often
found of a defined postsynthesis structure (average compo-
sition, particle form, surface orientation, steps) with a dynamic
component (segregation, surface-premelted layers, creeping
of carrier oxides onto metals, generation of substoichiometric
compounds) that is introduced through the reaction condi-
tions.
The general heterogeneous catalyst should be seen as
a mixture of static structures and dynamic components that
communally result in the active center. The synthesis of the
precursors and their chemical properties are as equally
important as the dynamics of the active phase. This does not
only apply to bulk catalysts, but also to supported catalysts
made of a “carrier phase” and an “active material”. This
partitioning of the structure of the catalyst into two pieces
after synthesis is functionally not sustainable because the
carrier usually has more than merely a “supporting function”,
a concept expressed through the term[27c, 29a, 246] “non-innocent
support”.[247]
In contrast, the discussion up until now shows that
a bulk catalyst, such as the systems for ammonia and
methanol or the well-known “M1” multielement oxide
catalyst,[23a, 201, 209f, 210b, 248] is also not functionally homogeneous,
but develops an active surface phase in a sort of “self- Figure 18. Electron microscopy images of active metal catalysts. The
view is of two-dimensional projections of columns composed of rows
support.” The processes leading to active centers then take
of atoms. The blurred regions are from structural elements (promo-
place in this surface phase. tors) not made up from the columns of the main components. A) Iron
Figure 18 shows some electron microscopy images of with promoters for ammonia synthesis, B) copper with ZnO for
prototype structures of metallic catalysts. Although clearly methanol synthesis. Image (A) shows an iron metal platelet with
defined structural elements can be seen, these images should a [010] orientation that contains many defects and is covered by
not lead to the tempting idea that active centers are visible islands of the promotor oxide (K, Ca, Al) that form a “crust” around
here. We do not know whether these structures actually carry the boundary surfaces. The copper particle has formed twin domains
and shows several facets and internal lattice defects. It is supported
out the reactions or how their structure under the conditions
and surrounded by ZnO. The very thin layer in the upper part of the
of electron microscopy is related to that under the conditions image is graphitic ZnO,[13a] while the mineral spacer in the lower
of the catalytic reaction. The images call for the methodical portion of the picture is mainly defect-rich ZnO with a zinc blende
development of electron microscopy toward the ability to structure. The structures were each identified by EELS and by analysis
produce such images under chemically defined conditions so of the Fourier transformations of the lattice. The samples were
that we will then arrive at a “chemical electron microscopy”. produced in a nanoreactor and loaded into the microscope under
We will now device the picture of a catalyst as being made- anaerobic conditions.
up from metallic central atoms that are bound into ligand
spheres[246, 249] by carrying out exchange reactions through
dynamic processes. The following statement applies respec- the stability of the active phase, but rather the elaborate
tively for both acid-base catalysts[250] and metal-free cata- structure of the active phase itself.[255] The chemical dynamics
lysts.[251] This picture further applies also to the most of such systems is only mentioned here with the descriptive
important case[211a, 252] of microporous and mesoporous acid- phrases “water vapor treatment” and “chemistry of pore
base catalysts (zeolites for example),[253] because the active openings”. Furthermore, metal-organic frameworks (MOFs)
layer makes up the surface layer of the of pore system. We with open network structures,[256] currently being the subject
begin to recognize this from the attempts[254] to unfold this of much study, also belong in this class of matrix-free
layer and investigate it in detail as a planar model. However, structure-stabilized active phases. Finally this picture also
in these systems it is not a bulk matrix that is responsible for applies to the group of Lewis acid-base catalysts.[94]

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3493
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

The ligands (counterions) are either present as intentional investigations. Examples for this approach can be found in
components or are generated from the reactants and the cooperative works from chemical physics and inorganic
catalyst precursors (typically OH). The mobility of the chemistry[27a, 46, 80c, 258] that aim to produce catalysts guided by
molecular system in solution with adjustable chemical poten- the construction of model systems. Of vast importance here
tial in an equilibrium of complex formation allows for the are combined theoretical and experimental investigations of
processes of chemical dynamics and leads naturally to cycles systems with realistic complexity[6, 30d, 191a, 236, 259] that illuminate
of the appearance and disappearance of active centers as an the reactive and spectroscopic characteristics that serve as
“active complex”. In the heterogeneous case, this “solution” landmarks in the examination of reactive systems. An
corresponds to a fluctuating termination layer on solids of example of this is the evidence of the passive termination of
approximately 1 nm thickness whose characteristics are metal oxides with M=O groups[162c, 191b, 260] that only allows
determined by the carrier phase. The termination layer is chemical reactions after the formation of defects in their
formed during the activation phase of the catalyst under the dense surface layer. Theoretical[13c, 261] and experimental
control of the chemical potential of the reactants. studies play similar roles for isolated clusters[47a, 194b, 262] and
If we imagine the same basic processes of ligand exchange their reactivity; they are able to give information about
and redox reactions in this termination layer, as are found in possible reactions from structures that may be active centers
molecular systems, we see that there is no significant differ- and also give indications about the charge state of these
ence between molecular and heterogeneous catalysis. The systems.
reason is that the actual reactions are not really as “hetero- We then still need analytical access to the atomic
geneous” as the standard model would lead us to believe. details of active surfaces under reaction condi-
Molecular or dissociated water is often produced in hetero- tions[13b, 24a, 33b, 51l,m, 108b, 125, 187a,b, 204b, 227, 263] without having to rely
geneous reactions and can play the role of a solvent, as can on models. The required methods are available to us today as
melting of the entire (or parts of) the active layer (for combinations of in situ analytical procedures. Table 5 shows
example, main group element oxides), or of additives that are some of the current common methods as well as the limits of
introduced as “promotors”. “Supported liquid phases” or their use.
“supported ionic liquids” are, therefore, no longer curiosities, This naturally cursory assessment of the experimental
but instead illustrate the normal way of how a catalyst ability of the different methods quickly reveals that despite
operates. However, the thickness of these layers differentiates the enormous arsenal, there is still a critical limitation in their
the systems, and with this we mean mainly layer thicknesses in applicability. The powerful methods for the analysis of
the monomolecular range and not optically visible layers. geometrical and electronic structures are generally not
This concept requires the abandonment of the idea of sensitive enough to the termination layer and its dynamics.
a rigid structure of active surfaces in heterogeneous catalysis The extremely surface-sensitive methods using chemical
described by translationally symmetric atomic arrangements reactions and adsorption with probe molecules are only
as observable in their non-active states. We sacrifice the conditionally suitable to define the investigated structures;
concept of analysis of an active structure through isolation in theory can be very helpful here to bridge this gap. For
a nonreactive environment and subsequent full physiochem- example, it can construct geometrical structures that are in
ical investigation. Instead we must demand that we will be agreement with observed vibrational frequencies. The meth-
able to analyze the geometric and electronic structure of the ods that are powerful for the characterization of structures are
termination layer well enough to develop structural and usually not sensitive enough to give information about the
functional quantifiable pictures of active centers without the termination layer. However, these methods can identify the
application of models. This will not be possible without transition of the termination phase into unwanted bulk
theoretical contributions achieved in such a way that the structures and, most of all, the preconditions for their
structural proposals are made using functional data and generation through characterization of the carrier structures.
spectroscopic findings rather than deriving these character- If thermochemical methods and time-resolved methods (on
istics from a predefined structure, as has been done up until the scale of the change of the chemical potential of the
this point. In this way we can help circumvent the dilemma of reactants, ms to s) are applied in tandem, a picture emerges
theory wanting to predict reactive non-equilibrium structures with options for the generation of the termination layer and
by concepts of total energy minimization. its dynamics. The case studies discussed here illustrate this
Model investigations receive a new function through this and the importance of the word “combination”, used above in
concept. They limit the number of options in the search for connection with the power of the analytical arsenal we in
active centers through the formulation of hypotheses and the principle have access to today, becomes clear. There is not one
exclusion of certain possibilities. Models of hypothetical overriding method but rather only a combination of methods
active centers can be designed[30d, 83a, 192] and their stoichio- with their individual abilities (in situ, structural sensitivity to
metric reactivity[257] as well as the reactivity of the related the termination layer, etc.) can be used to solve the problem.
nanostructures can be subsequently studied under reaction All of this also always remains connected with the analysis of
conditions. This does not allow for an unambiguous recon- the reaction, that is, with kinetics. We are, therefore, not
struction of the reaction, but it is possible in this way to define looking for the single “heroic” experiment that delivers
the stoichiometric elementary steps and significant elements ultimate insight but rather the synopsis of many critically
of the material chemistry of the active phase. Also, these can analyzed single experiments preformed under precisely
be used to set up boundary conditions for further in situ defined conditions. And these are all already available to us

3494 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

Table 5: Some of the approaches used today and their limitations.

Method In situ Investigated catalyst characteristic Sensitive to
capability termination layer
physisorption no surface, morphology yes
chemisoprtion, temperature- limited reactive centers, quantitative, qualitative yes
programmed desorption
microcalorimetry limited heat of sorption yes
vibrational spectroscopy yes structure and bonding geometry of Sorption phases conditional
optical spectroscopy yes electronic structure of active phases, adsorbate Phases very conditional
photoelectron spectroscopy yes quantitative surface analysis, depth profiles, adsorbate phases, surface electronic
conditional (UPS),
structure yes only as syn-
chrotron method
ion scattering no composition and structure of the outer layers yes
X-ray absorption yes electronic structure of active phases, adsorbate phases and geometrical structure very conditional
of active phases
scanning tunneling microscopy yes morphological structure and dynamics on mesoscopic scale no
X-ray microscopy yes morphology, chemical structure, tomographic, mesoscopic, analysis no
transmission electron micros- limited atomic structure of active phases, structural dynamics, chemical composition and conditional
copy electronic structure with atomic spacial resolution
X-ray scattering limited translationally symmetric geometric structure of bulk and surface (limited), real no
structure, defects
neutron scattering yes real structure and defects of the bulk phases no
inelastic scattering of X-rays and yes adsorbate phases under harsh reaction conditions (pressure, temperature) yes
neutrons
thermal analysis limited chemical dynamics, adsorbates, phase changes conditional
magnetic resonance limited structure and dynamics of active phases and adsorbates very conditional

now. This is what is meant by the statement that today we are until now not been able to furnish the practical catalysis
able[51j, 56b, 155e,h, 160b, 213, 264] to “look over the shoulder” of a work- researcher with a robust helping hand in unlocking the
ing catalyst without the aid of a model. material science side of catalysis. This deals with the question
It turns out that it is neither negligent nor ignorant to of how an optimized catalyst should be designed in terms of
remain with the original simplifications of the standard model composition, structure, and reaction dynamics. Especially the
and the resulting consequences. For many purposes this latter has not yet been investigated to a large extent because
makes for a perfectly adequate explanation of catalysis. If, a structural rigidity is usually assumed, the narrow limitations
however, the goal is to understand the fundamental nature of of which will now be discussed. Furthermore, the molecular
the phenomenon of catalysis or acquire the ability to “design” processes of activation and deactivation are only understood
catalysts, then this approach does not suffice and it becomes in exceptional situations[50b,f,j, 54a, 264a, 267] and can, therefore,
necessary to re-evaluate the simplifications leading to the seldom be treated conceptually.
standard model. Unfortunately, this has not been done at This analysis does not change fundamentally if the many
a desirable level and is the reason why progress remains slight academic reports on the discovery of new spectacular
apart from frequent announcements[5c, 7b,c, 37k, 265] and why, catalysts are included here. The synthesis of complex nano-
when the relationship between costs and output is considered, structured[268] material has given us a wide range of possible
the impression abounds that empirical catalysis research active materials.[211a, 44a, 269] Their function is, however, mostly
indeed leads to better results. We can remark that this is not not understood. They are usually not the result of previous
a product of our times. A similar conclusion can be found in functional considerations and the sustainability of their
other publications and in G. M. Schwabs summary of his effects under realistic conversion conditions[244a] is seldom
Handbook of Catalysis,[266] published in 1941. the subject of investigations. We are, therefore, in possession
of a rich supply of possibly interesting compounds that can
ignite the imagination to construct tailor-made systems.
13. How Did Material Science Arrive at Catalysis? Unfortunately, we do not have efficient methods to prioritize
these possibilities according to their effectiveness, other than
A fundamental criticism toward knowledge-oriented the laborious cycles of synthesis and functional analysis. A
catalysis research is its weakness in implementing functional more effective method would enable the selection of those
insight into new catalysts. The practical success in heteroge- systems whose exact functional characterization would lead
neous catalysis is due in most part to process technology, past their normal material and structural options in technical
which deals with the macroscopic dimensions of the links catalyst synthesis[5c, 118c, 211b, 270] to conceptually and theoret-
connecting the catalyst to reactor, as well as to the reactants, ically based conclusions. Regrettably, the effort involved
in the way described in Figures 1 and 16. For this, molecular proves too high a price, especially when it is considered that
events are largely insignificant. We only really understand this certain experiments[209g] will most likely not proceed success-
in a limited way with regards to chemical complexity and have fully.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3495
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

We are then charged with the task of proposing a path to

a catalytic material that begins by defining the desired
function. This is where physical-chemical-oriented catalyst
research can be put to use. In this field a concept can be
developed through the use of the standard model in the same
way that a combination of theory and model experiments[30d]
can lead to a knowledge-based approach to new systems.
Unfortunately, there have already been many broken prom-
ises of success along this road because the complexity of the
challenges was underestimated. In particular, only in rare
cases was the validity of the selected approach tested with
in situ analytical methods so that a fundamental understan-
ding[5c, 80c] of the observed function of the new catalyst could
be gained. The catalytic effect is only then a confirmation of
the validity of the physical-chemical concept.
This has led to the reservations about physical chemistry
in practical catalysis already discussed several times in this
Review. The author supports the view that in the cases where
empirical research no longer leads to an improvement of
a catalytic reaction, the remedy can be provided by the
concept of knowledge-based approaches sketched out herein.
If this decision were to be made early in the development
phase of a chemical reaction, with the knowledge-based Figure 19. Scheme indicating a knowledge-based approach to practical
catalysis. The arrows show the sequence of tasks that must be
approach experiencing repeated use, the associated costs on
completed multiple times to arrive at a sensible solution. The colors
a case-by-case basis would quickly be reduced. Exemplary of represent interdisciplinary cooperation that will be performed most
this would be the development of ammonia catalysts.[271] In effectively in a team setting. It is important that the steps toward
the history of this field, empirical research methods in the realization are accompanied from the beginning through process
early stages[131] have been combined with knowledge-based technology so that critical parameters of the procedure can be
approaches[24a, 97a, 127, 130] possible today and led to the expect- adjusted to the development of the process technology as soon as
possible. The common sequential handling of synthesis and testing is
ation[272] that even this system has a large potential for
unproductive because it only “calls on” the other disciplines when no
improvement if traditional reaction processes and active further immediate improvements in the performance of the reaction
materials are taken as a starting point to novel concepts of can be achieved. The continuous accompaniment of ecological consid-
ammonia synthesis. The level of understanding at this point, erations, which is always reacting to the changing eco-social condi-
suggests the optimum reaction processes possible with the tions and economic goals with respect to implementation, is not
current methods have already been achieved. shown.
A scheme of knowledge-based approaches is sketched
roughly in Figure 19. In the center is the elementary reaction,
often difficult to investigate, which we wish to be able to principle and the challenges presented by the complexity of
control. For reactions that are only to be accelerated, this reaching this goal can be met with tools already available
central reaction becomes a rate-determining step (“rds”) in today.
the nomenclature of Boudart.[101c] Unfortunately, there are no However, this appeal is neither original nor is it made for
shortcuts for a way through all of the stations. As soon as the first time here. The many efforts, for example, those made
a cycle has been completed and the knowledge gained about in the activation of methane,[47a, 95b, 274] may serve as good
the reaction solidified, parallel synthesis and test methods can examples. However, breakthrough success still has not been
be employed in a useful way in synthetic peripheral tasks.[273] achieved. Apparently the correct strategy has not been found
This is because the results must then only be sampled in the implementation of catalyst development using the
randomly as a check of whether the processes are still inside knowledge-based approach.
the bounds of the critical elementary reaction. We know today that many historical and current reports
However, the general case of a reaction with selectivity attest to the ability of a heterogeneous catalyst to adjust to
problems, such as water splitting, the selective hydrogenation reaction conditions. Here we speak of chemical dynam-
of CO/CO2 to oxygenates or to narrowly distributed hydro- ics[232, 246, 275] or of a “flexible”[133b] surface and realize that the
carbon mixtures, or even the selective oxidation of small quantitative characterization of the active centers is an
hydrocarbon materials,[59c] is an even larger challenge than enormous challenge. Without these centers we will hardly
ammonia synthesis. It is high time[42a, 97b] to tackle such be able to formulate a quantitative theory of heterogeneous
challenges in suitably organized projects. The progress made catalysis with predictive power. The situation in this field,
with homogeneous catalysis by carrying out many planned with such phenomenological complexity,[52a, 276] could be
reactions, at least in the laboratory, should provide incentives helped only by rigorous simplification to arrive at an
to attempt to reach a similar goal in heterogeneous catalysis. experimentally verifiable theory of catalysis starting from
The present Review aims to show that this is possible in the empirical-quantitative LHM.

3496 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

From this model, a theoretical connection can be made to

material characteristics[3a, 11, 82, 265e, 277] that a catalyst must
display on a molecular level. Therefore, a basis exists for
the design of a catalyst[7c, 14e, 23d, 82, 278] if the kinetic details of
a reaction are known as a function of the reaction conditions.
On the material science side, the electronic structure of the
catalyst must be understood in sufficient detail and with
chemical accuracy to describe the surface reactivity.
To use the knowledge of how a catalytic reaction proceeds
for the identification and optimization of suitable catalysts in
material chemistry, we contemplate the following concept,
whose derivation can be obtained in text books.[100]
At first we ignore the dynamic nature of catalysts by
separating the identification of a suitable catalyst to carry out
a reaction from the maximization of the number of active
centers per unit surface area. We also break down the choice
of material and its activation into different working steps, as
shown in Figure 19, which remain closely interlocked, a sit-
uation that can be assured through application of a functional
analysis of the material synthesis. Systematic findings about
the mode of generation and the reaction speeds of the active
centers are not accessible to us. In many cases this problem
does not become apparent because the critical reaction is the
activation of a small molecule on a step or another static high-
energy center (see, for example, Figures 7, 8, 12, 16, and 18).
The activation of hydrogen, nitrogen, oxygen, and water are
typical reactions of this kind. They take place at centers that
are generated as static defects after catalyst activation.
Different from these are the centers that mediate association
reactions and generate more complex molecules against the
trend of the reaction where thermodynamically stable prod-
ucts are formed. We know less about this, with the result that Figure 20. Development of the active phase on the catalyst MoVNb-
the empirical, or even theoretically supported, material TeOx in the reaction of propane with oxygen to acrylic acid. A) The
chemistry becomes more difficult. A typical field is the development of the relative cation composition in the bulk and in the
selective oxidation of alkanes. Currently only rules derived termination layer (photoemission at 620 eV kinetic energy) in different
from practical experience can be used as a guide and we reaction environments.[61] B) A selection of morphological data: the
perform work in this field based on the “seven pillars of needles of the crystal indicate the many steps along the (001) growth
axis. Long needles[210b] display a significantly higher selectivity (60 %)
oxidation catalysis”.[279]
with regard to the target product than do the short needles (5 %).
Figure 20 illustrates the challenge that appears with the C) The effect of water vapor leads to the development of a V5+ species
analysis of powerful oxidation catalysts.[280] The surface in the upper surface layers while the V4+ remains in the deeper layers
chemical composition is strongly dependent on the chemical and does not react with water. Over a reaction time of 10 h at 1 mbar
potential of the environment and varies much more then reaction pressure the proportion of the V5+ phase increases. At the
would be allowed from the quite flexible composition of the same time the yield of acrylic acid improves. However, the elemental
unit cell structure of the oxide[281] alone: this must be a case of concentration in the segregated termination layer remains unchanged.
chemically induced segregation.[282] In addition to the heavy
Te segregation, a vanadium phase consisting of V5+ species
also forms in the presence of water vapor. The existence of formally exclusive presence of the V4+ species[51d, 287] existing
this phase is directly related to the target reaction[61, 282b] and, in the VPO structures, the active phase under reaction
interestingly, no changes in the chemical composition of the conditions at 1000 mbar pressure is comprised of V5+ [194a, 288]
surface take place during the formation of the phase. The formed by the chemical dynamics of the activation in hot
continuous presence of V4+ amongst the Nb5+ indicates that reaction gases. This is an example of an especially difficult
the bulk structure remains completely stable, in agreement form of the pressure gap, because this information was not
with in situ X-ray structural analysis.[201, 283] The chemical obtainable at the normal 1 mbar reaction pressure of the
dynamic in this system remains limited to the uppermost in situ analysis with surface analytical methods.[187c, 194a, 289]
termination layer.[60b] In the high-performance system However, the participation of the V=O bond at the active
VPO[204b, 222, 284] (vanadium pyrophosphate) this is not the centers was correctly identified in these experiments and is
case. Instead, the generation of the active phase[199a, 285] is also supported by earlier studies.[290] High-pressure methods
connected to a complex change of the bulk structure.[286] This using RAMAN spectroscopy[291] have turned out to be not
complexity long prevented the recognition that in spite of the surface-sensitive enough to detect the active phase whose

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3497
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

thickness is only about 1 % of the wavelength of the light carbons, minimizes the reductive effect of hydrogen, and
used. However, using EXAFS[288] and chemical probes[292] the leads to the optimization of the desorption of the target
V5+ species was also identified as the catalytically relevant product (site isolation).[14c] Secondly, the generation of sub-
species. surface hydrogen must be prevented because this activated
After we have convinced ourselves that the many relevant form of hydrogen, together with Pd, creates a hydrogenation
catalysts of today would be difficult to treat with a knowledge- catalyst[181c] that is unselective[196, 218] in this reaction. In this
based approach using our current knowhow, we may turn to last study there is classic evidence for the validity of the LHM
a more simple case. We will look at the selective hydro- in the case of hydrogenation.
genation of CC bonds (alkynes) to C=C bonds (olefins) over With this concept it was possible to develop intermetallic
metal catalysts. From Equation (25) and its interpretation, we compounds[300] into catalysts. The line phases PdGa and
know that the most essential value for the reaction rate is the Pd2Ga are both able to keep hydrogen out of their lattice
surface coverage by reactants. When atomic hydrogen is often structure[301] and possess terminations[300d, 302] with structures
found near adsorbed alkynes a high reaction rate[293] will that completely, or nearly completely separate the Pd atoms
result. As the bonds between alkynes and a metal are from one another by forming cagelike motives from Ga and
significantly stronger than between a metal and alkenes[294] Pd. The strong covalent contributions of the bonds necessary
it can be assumed that there is a finite chance it will desorb for this special structure also shift the Pd d band to
before it is further hydrogenated by hydrogen to an alkane. significantly lower energies and effect a modification of the
There is an established reaction mechanism[10a] and model electronic structure. The concept is successful conceptually as
experiments[147, 181c, 295] that exist for this. The reactions (28) well as in reality, for example, with supported nanoparticles,
and has led to stable and effective catalysts.[225, 226b]
2 C2 H2 þ H2 ! 2 C2 H4 ð28Þ It turns out when considering the postulated control of the
causal relationship between catalyst success and original
and (29) concept that this is indeed comprehensible[225, 226c, 303, 305] for
annealed single crystals and pure samples free of surface
2 C2 H4 þ H2 ! 2 C2 H6 ð29Þ oxides.[304] Real nanostructured catalysts and unannealed,
crushed bulk samples show, however, another working
have, furthermore, been studied intensively on a theoretical principle: the generation of a Pd nanostructure protected
basis[178b, 296] and serve as case studies for a knowledge-based from sintering by a Ga2O3 nanoparticle was observed[226b, 306]
development of proposed new catalysts. In one theoretical as long as it was involved in an acetylene hydrogenation
approach a method was introduced[297] where a non-noble reaction. The high selectivity was achieved by Pd-C subsur-
metal catalyst could be identified for the reaction of the face compounds.[196, 225, 299a] These can also cause pure Pd
selective hydrogenation of acetylene. Normally reaction (28) nanoparticles to be highly selective if suitable structuring of
is used selectively to purge an ethylene stream of contami- the Pd precursor is available to promote their growth along
nation by acetylene and is accomplished with Pd-Ag with carbon supports for stabilization.[264f] Properly function-
alloys.[189b, 296, 298] The problem is to avoid total hydrogenation ing Pd2Ga nanoparticles[307] with a stable intermetallic struc-
(29), which destroys the starting material for the polymeri- ture can also be produced and utilized[225, 308] on nanocarbon
zation target reaction. With the help of a model, discussed supports. However, if more complex alkynes are to be
below, a series of alloys with base metals was proposed that hydrogenated, the concept does not work well[56c,d] for two
can carry out very selective hydrogenation. The data of reasons. First, the complexing effect of the substrates corrodes
experimental reproduction are very convincing.[297] However, the surface and, second, the reactivity of the pristine
the products from the synthesis of the catalyst were not intermetallic compound cannot compete with the surface-
checked to determine whether the predicted electronic modified Pd systems (Lindlar catalysis) when it comes to
structure of the alloys was in fact responsible for the observed substrates that are difficult to hydrogenate.[56d, 309]
effect. There is an array of possible explanations for the several highlights of this development are shown in
measurements, the proffered theory[297] being only one of Figure 21.[300a] By using in situ X-ray diffraction, the gener-
these. Alternatives are based on morphological variations[162a] ation of PdH was established on nanoparticles supported by
of the base metals through carrier interactions or selective nanocarbon[310] along with PdCx after the reaction. The
poisoning of steps on the base metal surfaces, on reactions of concept of the modification of Pd surfaces by the generation
a component of an alloy with the carrier, or on a selective of intermetallic compounds can be studied on single crystals
capping of the base metal with the alloy component. As these as well as on polycrystalline systems by determination of the
alternatives cannot be excluded, the experiment is not fully binding energy of the CO probe molecule. In addition to
able to verify the theory. showing how drastically the absorption characteristics (sur-
These reservations are a consequence of the experiences face coverage) change when Pd metal is converted into an
with the knowledge-based approach from Figure 17. The intermetallic compound, Figure 21 B also shows the concept
standard model and the specific model experiments that lead of site isolation. However, this is only the case when
to the theoretical approach sketched above predict that the tempering is used to achieve a surface termination[304] in
desired reaction can only proceed under two conditions: first, which Pd is actually surrounded by Ga. The shift of the Pd
the active center cannot consist of Pd terraces[294] but rather of d band through the generation of the intermetallic com-
isolated Pd atoms. This prevents the dissociation of hydro- pounds can be easily demonstrated with electron spectrosco-

3498 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

them so strongly to the carbon that the do

not separate even under hydrogenation
conditions.
These observations show that it cannot
be assumed, even with seemingly straight-
forward reactions, that the verification of
a concepts function within the framework
of a specific model system can be directly
transferred to another system. Specifically,
from this example we can learn the
following lessons, which are also found in
a similar form in other fields of nano-
catalysis[27c, 311] with metals.
1. The model experiment of a comparison
of a Pd nanoparticle with a Pd single
crystal showed the importance of the
structuring of one and the same mate-
rial for catalysis.
2. The concept that catalysis is only
determined through surface processes
must be corrected: subsurface regions
of the catalyst can be of decisive
relevance if they are brought into the
reaction process at critical chemical
potential thresholds.
3. The LHM of the standard model was
successfully applied to a complex reac-
tion beyond ammonia synthesis and
CO oxidation. This forms a solid basis
for the general concept of heteroge-
neous catalytic reactions.
4. Model experiments whose results lead
to physically resilient concepts can be
used to plan new approaches to syn-
thesis that would likely not have been
possible with the planning of mere
empirical experiments.
5. The choice of complex materials such
as an intermetallic compound can sig-
nificantly simplify a specific catalytic
application. With a good knowledge of
Figure 21. Several observations[300a] concerning the use of A) Pd, B) PdGa, and C,D) Pd2Ga as the material characteristics a catalytic
a catalyst for the selective hydrogenation of C2H2 to C2H4. A) In situ X-ray powder data concept can be accurately imple-
(detailed region of the Pd(111) reflections) of approximately 4 nm Pd particles on carbon mented.
nanorods: a) fresh catalyst in He at 300 K, b) in 4 % hydrogen at 400 K, c) in 2 % acetylene, 6. Catalysts are seldom “universal”: the
4 % hydrogen at 400 K, d) as in (c) at 503 K, e) in nitrogen at 503 K. The blue lines show transfer of solutions from one reaction
reflection positions of pure Pd, the red lines the shift for b-PdH, the intermediate reflections
to another is only conceivable on an
show the generation of PdCx.[297] B) Temperature-programed desorption of CO on single
crystals of Pd and Pt with a (111) orientation and PdGa with two terminations. C) XPS from
abstract level and as a first approxima-
Pd and Pd2Ga nanoparticles on carbon nanotubes: a) Pd in 0.2 mbar He, b) Pd2Ga in tion.
0.2 mbar He, c) Pd2Ga in 0.2 mbar H2. No hydride phase is formed for the intermetallic
compound. D) TEM and EDX analysis of Pd2Ga on carbon nanotubes. The TEM image shows Here it is worthwhile to dwell a bit
the good adhesion of the Pd2Ga and the EDX analysis the homogeneity of the elemental longer on the topic of material design for
distribution.[225] catalysts because we may be able to better
understand the ideal “catalyst design”. We
can take from the standard model that the
py along with the stability against hydrogen.[299a] Electron degree to which the surface is covered by reactants is
microscopy can be used to show that the nanoparticles are of determined by the material flux and the temperature
similar and homogeneous composition. Their structure cor- (together in a reactor with a local chemical potential, see
responds to the bulk phase and the intense interaction binds Figures 15 and 16) as well as by a material constant that we

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3499
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

call the sticking coefficient. This factor contains the details of observed active materials carry out the reaction using the
the chemical interaction between the reactants and the same mechanism (often difficult to verify) then we can exploit
catalyst. The details of the chemical bonds of interest here a general relationship in physical chemistry between the
are only included implicitly because the derivation of this (thermodynamic) strength of a bond between a catalyst and
coefficient is based on the kinetic gas theory[100] of corpuscular reactant and the height of the energy barrier (kinetic) that
interactions and not on explicit chemical bonds. If the sticking must be overcome to complete the reaction. This relation-
coefficient is written in an Arrhenius form ship[3a, 313] states that the barrier and the binding energy are
correlated with a schematic representation found in
DEakt
S ¼ S0 e kT ð30Þ Figure 22.

then we may recognize after looking at the value of the

activation energy which kind of reactant adsorption we have:
there can be weak binding of molecules (0–20 kJ mol1) or
strongly activated, dissociative adsorption (40–200 kJ mol1).
The degree of coverage, which is of interesting here, is
then given by Equation (31)

qðtÞ ¼ 1  eFS0 ðTÞt ð31Þ

Here, F is the reactant flux on the surface of the catalyst

and t is the time. To eliminate this we again use the
approximation of the stationary state and recognize that the
sticking coefficient and its temperature dependence deter-
mine the coverage. The local chemical potential and the
chemical nature of the reactants are built into the parameters
of the sticking coefficient S0.
In many cases[263] it is favorable if the surface coverage
from different reactants is approximately the same, and is low.
Then there is a high probability that the reactants will
encounter one another at one of the rare active centers. If the
reactants are not too chemically different, a single active
phase can fulfill this requirement if the parameters chemical
composition, pressure, and temperature are correctly chosen
and a highly active catalyst results. It will be much more
demanding to choose a material if the reactants are chemi-
cally strongly different or if subsequent reaction cascades are
necessary for the generation of a product. A “multifunc-
tional” catalyst can be envisioned with one redox function and
one acidic function. In contrast to a homogeneous system in Figure 22. Relationship between the barrier height for the activation of
which multifunctionality is difficult to integrate into a single a reactant and the strength of its interaction with a series of active
active complex, multifunctionality at the surface leads to phases (above) and the reaction rate of the corresponding catalytic
conversion (below). The relation is shown for two reactants (red,
fewer problems because chemically different groups can co-
black). The red field denotes the regions of the material descriptor
exist in proximity without influencing each other through (Dmat) in which the reactant is bonded too strongly to the catalyst, the
directional bonds. green region shows where the interaction is too weak. Erea is the
We can choose a middle-of-the-road case with two enthalpy of reaction of the rate-determining process, EA is its activation
reactants that are fairly similar. Then, we may assume that barrier.
we have a descriptor available that allows us to quantitatively
describe a family of possible materials. Such a descriptor
could be the strength of the interaction between the active We now have the necessary tools to choose materials for
phase and an activated reactant. An example could be a metal reactions with a known mechanism (at least with a known
and a hydrogenation reaction. The descriptor would then be rate-determining step). Equations (30) and (31), as well as the
the bond strength between hydrogen atoms and the active discussion on the dynamics of the catalyst, show, furthermore,
phase. Alternatively, we could take the descriptor to be the that this choice is not a constant, but is rather dependent on
bond strength of the molecules that are to be hydrogenated. the reaction conditions. For this reason the search for
As a further approximation, the reaction will have only one materials without a parallel consideration of the reaction
rate-determining step with one corresponding species. This is conditions is not recommended. From this we see again the
often the case for a hydrogenation, although not for an origin of the cyclic operation of the knowledge-based
oxidation, if we consider the dehydrogenation and oxygen approach in Figure 19 and postulate that comparing a series
transfer to be substeps.[152a, 228b, 312] If we suppose that all of the of materials under the same test conditions is not sufficient.

3500 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

Only undertaking a very coarse search for large differences in scaling relation applies to several structures. However, if the
the reaction rates (logarithmic scale in Figure 22) will enable energy resolution is improved in these relations, it is found
the combination of the important factors into a single that they congregate into groups. Hydrogen stands alone or as
constant on the first go around. a bonding atom of a fragment, it lies far from the universal
This qualitative approach, known in the literature as the relation, and defines its own scaling relation. The prediction
“Sabatier Principle”,[82, 313a, 314] will be useful first when we of catalysts for whole reaction networks, is, therefore,
have numerical values for the descriptor. The measurement of reduced[320, 323] to solving linear equations with numerical
EA or ERea is essentially possible with calorimetry on real values from quantum chemical calculations. These can be
systems or on model systems[148a] with well-defined elemen- stored in a database and used to create new combinations of
tary steps. However, this is so costly[147, 149, 150, 151] that attempt- fragments and materials.[324] Sets of curves or “volcano
ing to find materials on an empirical basis is preferred. plots”[9a, 277b] are generated with forms similar to that found
Here theory has made a novel approach possible. This is in Figure 22: the minimum curve limits the possible perfor-
due to the now high quality of the description of the surface mance of the catalyst from that particular set.
electronic structure of chemically complex structures. Based If this predicted performance is to be improved, it follows
on the enormous progress in the theory of molecular from the theory[272] that the validity of the scaling relations
processes in catalysis,[3a, 9a, 11, 278c, 315] we are now, with a few must be rendered untrue. Such a situation will only be found
curtailments with respect to precision, able to calculate in a set of compounds if it is possible to decouple the
enough numerical values to get a solid idea about many activation and reaction of reactants from one another and
compounds and a large number of relevant reactions. With from the local electronic structure of the catalyst. This can be
this information we can make predictions about specific achieved if a modification of the mode of operation of
materials, their most advantageous surface structure, and the catalysts takes place that is not found in the models defining
maximum achievable reaction rate. For the test case of the scaling relation. Some of these are dimensionality of the
ammonia synthesis[101b, 130a, 316] this was done extensively up to active phase (exact structure, bulk, cluster, thin layers), the
the point of proposing new binary metal compounds[3a, 265e, 277b] spatial morphology of the system (confinement effects), the
with substantial catalytic performance. This achievement was dynamics of active center formation (static active centers are
made possible due to the details of the dissociative adsorption standard), the reaction environment in which the active
of nitrogen on the catalyst being theoretically[96,1 112b, 317] and material is placed, or the use of nonthermal energy sources
experimentally[22d, 92b, 130b, 229, 318] very well understood, as are mentioned above. These possibilities result from the consid-
the details of the rate-determining step. The work on the eration of Figure 16 that describes the recursive coupling of
ammonia system has advanced to a degree that there are the entire system “catalytic process” with the active centers.
theoretical predictions stating just how much further this The methodology of predictions from scaling relations has
reaction can be developed.[272] These appear rather utopic at been applied to diverse and complex systems including the
this point if the history of the actual development[130b, 131, 271a] of problem of the hydrogenation of acetylene.[296] The result
this reaction is regarded in comparison. In theoretical work, were numerous suggestions[297] for new systems, one of which
however, a very important clue can be found to the method covered the intermetallic compound PdGa just discussed.
leading to such breakthrough results. This approach was also able to reproduce the verification that
For this we return to Figure 22. It shows that many the selectivity of this reaction is controlled mainly[196, 325] by
relevant adsorbates, which may act as possible reactants, have the formation of Pd subsurface carbon bonds. The daring
a linear dependence between their heat of formation, and presentation of the correlation[297] between selective hydro-
their position in the periodic table of the elements. This is not genation ability and the monetary value of the catalyst was
an accident,[82] but rather it can be explained with C, O, and even risked to predict cost–effect relationships.
N atoms using as the bonding mechanism[101b,319] the inter- A similar and fruitful analysis was undertaken in electro-
action of molecular fragments[320] with the d electrons catalysis.[14e, 44a, 278c, 326] Under the assumption that the catalysts
(d band). This is mainly responsible for the bonds in transition were of a metallic nature, verifiable predictions were made
elements that are especially relevant in catalysis. The regular about the composition and structure of electroreduction
shift in the position of the core of the d band,[230a] as well as the catalysts (ORRs) for oxygen (see Figure 2). Further exper-
band edge, with position in the periodic table (filling of the imental support has also been found.[322a, 327] The development
d shell) explains the change in interaction energies. If this of electrocatalysts for hydrogen evolution[278f] was also
somewhat coarse analysis is refined with input from studied with this method by using a selection of over 700
alloy formation,[7c, 178b, 277c, 278f, 297, 300c, 315c, 321] steps on the sur- systems. Notable in this work is the theoretical analysis that
face[24a, 96, 132c] and strain in the unit cell[322] of the metallic also included the stability of the phases against decomposi-
catalyst, many new elements are brought into the previously tion. An “unexpected” prediction was made of an effective
monotone shift of the d band. If this “scaling relation” is material (a Pt compound) and subsequently experimentally
investigated to find a series of catalytically relevant fragments verified. The oxidation reaction of oxygen (OER) was studied
over a large number of elements and structures on a catalytic on oxides[44, 328] and compared mechanistically with the water
surface, it can be seen[14e, 82] that the scaling relations have splitting found in nature.[44b] The result was the explanation
similar trends and that the exact structure of the adsorbates for why no better catalyst than RuO2 can be found for this
has no significant influence on the results. This leads to reaction although many different systems have been sug-
a certain “universality” in catalysis because one and the same gested. And this despite the significant overvoltage needed to

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3501
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

successfully achieve the reaction with RuO2 in comparison to “concepts” based on faulty physical reasoning that are
the thermodynamic limit of the barrier. Unfortunately miss- vehemently defended. This does not bring us any closer to
ing from this work is information about the nanostructuring our goal. The method of an aesthetic-artistic design of active
and stability[195d, 329] of the system, characteristics that present centers can act as an enrichment of the necessary creativity
large experimental[51j, 329a] challenges. but is no replacement for the targeted construction based on
The theoretical design of a catalyst requires the subse- chemically correct models of the reaction and free from
quent synthesis of the material in question for its verification material classes and structural guidelines.[5a,b, 265g] We know the
as well as knowledge of the composition on mesoscopic and way and possess the basic physical principles but are
macroscopic scales. This material then needs a functional continuously tempted by the complexity of the kinetics and
characterization to show that the theoretical predictions material chemistry of heterogeneous catalysts to take short-
about the active centers are in fact borne out. This has been cuts. This results in a mesh of simplifications that impede
done in several cases[7c, 265e, 278a,f] if, perhaps, in still a somewhat rather than help the transfer of new knowledge to practical
unrefined manner. However, side effects of the reactivity catalytic situations.
appeared that influenced the stability of the catalyst in an The theoretical approach discussed up until now selects
adverse way and no technical realization has been reported. materials according to their function based on simplified
There are reservations about continuing along this path of assumptions of structure and morphology and on neglecting
theoretically supported design of catalysts with the current the influence of reaction conditions on the catalyst. This
lack of chemical precision.[14e, 82] If the spectrum of the current corresponds to the description along the abscissa in Figure 22.
reports of this approach in the literature is considered, it can To make a contribution to the material science of catalysis
be quickly seen that it is not the aim of this theory to provide with theory, the orthogonal path can also be traversed that
a highly accurate prediction for one good catalyst: the rates runs along the ordinate. In this case we obtain precise and
are given on logarithmic scales over several orders of detailed information about the predicted course of the rate-
magnitude. In contrast, in real cases we are interested in the determining step and about the structural details of a chosen
effect of a catalyst within a maximum factor of 10, usually, catalyst under reaction conditions. From this we can learn
though, within a factor of two. Therefore, this theory of much about the constitution and dynamics of a given catalyst.
scaling relations may be regarded as a selector that supplies us It necessitates an elaborate theoretical apparatus with
with material suggestions on the scale of descriptors in sufficient “chemical” accuracy that allows quantum mechan-
Figure 22 (approx. in the middle) in which we can optimis- ical calculations for adequately complex models. Further-
tically search for solutions. This approach excludes possibil- more, a description of thermodynamic functions at finite
ities that are not relevant as possible solutions and narrows temperatures and pressures is necessary to construct phase
the space of material options for potential hits. At this time diagrams of the catalyst as a function of the chemical
the theory utilizes rough structural models, cares hardly for potential of the reactants. This theory[9a, 11b, 152b, 235, 330] empha-
size effects and not at all for chemically complex compound sizes conceptually the idea of the unity of catalyst and
catalysts. The author supports the view that all methods reaction, as described in Figures 12, 16, and 17.
enabling us to distance ourselves from empirical development The path to deciphering the systematic behavior of
are important for further progress in catalysis research if we a catalyst step-by-step with the help of theory seems to be
can learn from the observed differences between theoretical very promising at this time.[2g, 13c, 47a, 94, 331] The approach has
and experimental results. There is no one single factor that been applied successfully to, among other procedures, hydro-
fundamentally impedes improved precision. One important genation catalysis over noble metals in a very realistic
point is to use helpful information from models to exclude way.[10a,b, 14b,c, 265c, 296, 313b, 332] For methane synthesis a reaction
options that will certainly not lead to the goals we have set. network was also calculated extensively.[146] In these cases,
The standard for judging on theory-supported catalysis however, rigid assumptions about the nature of the catalyst
research should not be an immediate technically useful were made. For the selective oxidation of propane,[152a]
result, or even time-saving compared to a purely empirical a reaction network based on qualitative suggestions[312] was
approach, but rather the sum of the physics-based conceptual quantified. Here the splitting of the selectivity into partial and
and methodological insight that was gained throughout the total oxidation was described as a function of the redox state
process. The value of such findings results when they are of supposed active centers. The result is that a quantitative
utilized in a knowledge-based approach, such as suggested in and experimentally observable[333] descriptor (surface cover-
Figure 19. The time saving will materialize after we have age from peroxo species) can be theoretically established for
applied this concept to several problems because we can understanding the principles of selectivity control. The
always use the knowledge-based insight (“why has something resulting development shows how the fields of knowledge of
been observed”) again and in a new context. This is not molecular and solid-state catalysis can mutually benefit one
possible in empirical strategies with synthesis tests only another, especially if they become conceptually connected by
because we have no knowledge of the causality of results in the bridge of theory.
new situations. A further general and instructive example of the
The most promising route to move catalysis from an success of this theoretical approach is the investigation of
empirical science to a design science is the closing of the CO oxidation over Ru. Starting from xtensive experi-
knowledge gaps that still abound. A sadly all too common mental investigations[334] a comprehensive theoretical analy-
denial of its existence shows itself though the proposal of sis[10c, 11a, 278b, 335] of the reaction was undertaken. One out-

3502 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

standing result for material chemistry is that it became clear

with the help of the theoretical approach that the highest
reactivity of the system is achieved when the surface does not
form any single stable structure. Instead, the surface should
lie energetically exactly in the middle between the oxide and
metal. This state can be interpreted as a surface oxide and/or
a defect state in an oxide film.[156b, 264b] It clarifies in part the
contradictory perspectives[334d] on the nature of the active
phase and on the oscillations of this reaction[336] that have
been found. Clarification is also provided for the result that in
the oxidation of methanol over Ru systems the most active
state is a surface oxide that, finally adjusting to stationary
conditions,[220, 337] forms either from a metal or an oxide
precursor.
A further example of the possibilities of theory[9b, 11b] to
help interpret complex experimental results is given by the
silver/oxygen system used in catalysis for the epoxidation of
ethylene and for the partial oxidation of methanol to
formaldehyde under different conditions. In this case the
results led to a new understanding about the dynamic state of
catalysts. Merely the fact that the reactions (32) and (33):

2 C2 H4 þ O2 ! 2 C2 H4 O ð32Þ

2 CH3 OH þ O2 ! 2 CH2 O þ 2 H2 O ð33Þ

lead to two very different products very likely means that

silver is able to form different reactive forms from active
oxygen. This happens under conditions at which the possi-
bility of a metal oxide phase change can be excluded because
of the instability of silver oxide. The system has been
extensively studied experimentally.[126b,179j,k, 187e, 338] Several sig-
nificant results are summarized in Figure 23.
Under high-vacuum conditions, the system forms a series
of reconstructed phases that represent ionic silver strongly
interacting with atomic oxygen derived from the dissociation
of molecular oxygen below room temperature. As a limit,
silver oxide surface phases form up to about 473 K. Above Figure 23. Several experimental results for silver as an oxidation
catalyst. A) Stages of the interaction of oxygen with silver beginning at
this temperature more weakly bound forms of atomic oxygen
ca. 423 K. a) Adsorption and reconstruction, b) surface oxide and
exist along with oxygen dissolved under the surface. Above facilitated defect formation, c) as in (b) plus electrophilic oxygen
773 K the latter segregates into a form of strongly bound through subsurface oxygen integration, d) surface oxide. The process
oxygen localized in the surface. The bulk phase can hold of integration of oxygen into the bulk without the generation of an
a large amount of oxygen and even acts as a membrane at high oxide proceeds much further at ca. 723 K, although only a small
temperatures for the removal of oxygen from air. The amount of adsorbed oxygen is available on the surface. B) XP spectra
(O1s) of silver in 0.5 mbar oxygen: bottom: 2 h at 423 K, middle:
existence of oxygen dissolved in the metal bulk has been
30 min at 723 K, top: 5 h at 723 K. C) XPS (O1s) desorption series of
verified many times.[179k, 340] In the example experiment shown oxygen from the state (c) in (B). D) XPS (O1s) from silver particles
in Figure 23 C, oxygen originates from its diffusion deep in the (100 nm) in oxygen and ethylene (1:2, 0.5 mbar) at 503 K for the given
bulk where, after depletion of the subsurface region, new times. E) Correlation of the proportions of the quantities of electro-
oxygen from the bulk appears through desorption and causes philic (magenta and dark green in (D)) to nucleophilic (red, light
a second desorption maximum. The existence of several green in (D)) oxygen versus the in situ observed selectivity to ethylene
different species of atomic oxygen can be seen very clearly in oxide. From Refs. [21b, 339].
the measurements in Figure 23 as well as in related studies in
the literature. They can be separated into nucleophilic and
electrophilic forms. Nucleophilic forms are stable, cause the alkoxides. The combination of a subsurface layer of oxygen
adsorption of organic substrates, and dehydrogenate hydro- with adsorbed oxygen directly above is especially effective at
carbons. Electrophilic species only result from the interaction generating very electrophilic species (see Figure 23 A).
of silver with subsurface species. They show only a slight The electronic structure theory of this system was used to
stability, tend to diffuse into the bulk, and cause the transfer construct a phase diagram for the oxygen/silver system for
of oxygen onto organic, activated substrates such as olefins or catalytically relevant conditions. The diagram shows the

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3503
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

6 AgOp44 ! Agð111Þ þ 3 O2 ð34fÞ

fundamental trends of the structural development in a quan-
titative way.[341] It is problematic for the theory to represent 6 Agstep ! Agð111Þ ð34gÞ
the bulk species, identified as energetically unfavorable for
many states of the system, in comparison to segregation. This AgOsub ! Odis þ Agsub ð34hÞ
is in concurrence with the observation that these states
emerge under dynamic oxygen supplies and sufficient tem- AgOsub þ C2 H4 Ð ðAgOsub ÞC2 H4 * ð34iÞ
peratures as well as with the application of nanostructured
ðAgOsub ÞC2 H4 * þ ðAgOsub ÞOel ! C2 H4 OEO þ 2 AgOsub ð34jÞ
samples. They are metastable against segregation if they are
to be isolated under normal conditions. The state in Fig- ðAgOsub ÞC2 H4 * þ AgOnu * ! C2 H4 OAA þ AgOsub þ Ag ð34kÞ
ure 23 A,d was, however, found to be energetically equivalent
to a state similar to that in Figure 23 A,b. Reconstruction and C2 H4 OAA þ Ag Ð AgC2 H4 OAA * ð34lÞ
defect formation via oxygen adsorbate phases on silver were
found to be in wonderful agreement with experimental results AgC2 H4 OAA * þ 4 AgOnu * þ 2 ðAgOsub ÞOel !
ð34mÞ
from theory.[339] Furthermore, it was shown that subsurface 2 CO2 þ 2 H2 O þ 5 Ag þ 2 ðAgOsub Þ
oxygen can even promote the process of defect generation as
a starting point for restructuring and is supported by earlier Besides adsorbed molecular oxygen (AgO2*) we find
experiments.[342] In a very early theoretical approach the atomic oxygen as nucleophilic species (AgOnu*), electrophilic
formation of electrophilic species by the interaction of surface species ((AgOsub)Oel), surface oxide (AgOp44), and subsur-
oxygen with subsurface oxygen was found to be significant for face species (AgOsub) in agreement with the results presented
the generation of ethylene oxide.[343] This is of interest in Figure 23.
because the proposal was made before the experimentally The set of Equations (34) describes adsorption (a,b),
water-tight proof of electrophilic oxygen[344] and the dynamics formation of active oxygen species (c–e) plus the loss of active
of the silver surface were obtained. The presence of adsorbed sites and restructuring of silver (f–h), in addition to the
oxygen facilitates defect formation in metal, enables exten- selective (i,j) and nonselective (k–m) oxidation of ethene. The
sive structural dynamics, and leads to reconstructions of the overall combustion reaction is formulated in a rather cursory
surface at mild temperatures. This, on the other hand, has manner, as many details of this process are not well known.
effects on the formation[339] of nucleophilic and electrophilic These equations may serve the purpose of understanding how
adsorbate phases. many properties of the reactivity of the catalyst silver with the
The reconstruction behavior can, furthermore, be con- reactant oxygen except its dissociative activation need to be
trolled through the formation of alloys. The system Ag-Cu has identified and quantified to arrive at a meaningful description
been reported experimentally[333] as being significantly more of the target reaction according to the central scheme in
selective in ethylene epoxidation than pure silver. This was Figure 16. We have not yet reached that stage of spending the
then attributed to the formation of an alloy. In a later effort to quantify the reaction parameters of catalyst forma-
extensive theoretical study[215] it was shown, in conjunction tion despite that fact that we could do this from our level of
with parallel in-situ photoemission measurements, that it was analytical capabilities.
in fact not an alloy but rather the formation of a thin surface Parallel to the progress with theory for catalysis, the
phase of CuO on reconstructed silver. The exact structure and chemistry of the synthesis of catalysts must also be further
morphology of the silver was found to be sensitive to the developed.[145] Today we see the refinement and analytical
chemical potential of the oxygen and a corresponding state substantiation of conventional synthesis procedures. This has
diagram was developed. Several proposals[334] for active enabled important progress mainly in the homogeneity and
structures have resulted from this, however, although they scalability of catalytic materials with a considerable deviation
are built on a theoretical basis, they were made with from thermodynamic stability leading to good catalytic
consideration of Cu and the structural dynamics of silver. properties.[109, 209a,c,h, 347] In addition, there is much activity in
We can use the example of silver-catalyzed epoxidation of the synthesis of novel materials[268d, 348] whose potential for use
ethene to show that a quantitative description of the kinetics as catalysts is being investigated. This very creative procedure
of a catalytic reaction should also contain reaction steps for is, however, less compatible with the planned advancement
creating and losing active sites. The interaction of reactants we have attempted to motivate here with theory that begins
with the silver catalyst should thus be specified more in with a targeted property profile.[244a, 278a, 322a, 327a, 349]
chemical detail than just using the asterisk symbol for the The high specificity of their catalytic effect, which is
formation of an adsorbed species [Eq. (34)]. connected to defects in the bulk structure, generally puts high
demands on the synthesis of heterogeneous catalysts. It is not
Ag þ O2 Ð AgO2 * ð34aÞ sufficient to only produce a single phase and then test its
catalytic activity; along with this a suitable surface structure
AgO2 * þ Agstep ! 2 AgOnu * þ Ag ð34bÞ
must also be found. This importance of the specificity of the
AgOnu * þ Agsub Ð AgOsub þ Agstep ð34cÞ active catalytic structure,[121] given by the sensitivity of the
catalytic reaction to the local electronic structure, is often
AgOnu * þ AgOsub ! ðAgOsub ÞOel þ Agstep ð34dÞ underestimated. This leads to the view that the synthesis of
catalysts is a magical assignment because the usual analytical
AgOnu * þ Ag Ð AgOp44 þ Agstep ð34eÞ procedures often cannot differentiate well enough between

3504 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

the local surface structures. The result is that no structural s.[62c, 211a, 256a, 356] Furthermore, the chemical dynamics of the
“difference” can be found between substances with different material under the planned reaction conditions must also be
catalytic properties that are chemically otherwise identical. planned and adjusted. Then nanostructuring must be chosen
It should be mentioned at this stage that the multiple such that it affects the activation kinetics in a favorable way.
approaches for the selection of catalysts with combinatorial Here, stationary active centers must also be provided through
synthesis[273a, 350] that the use of chemical composition as the the surface structure (“roughening” itself may not be
single variable will not, with any surety, result in the expected sufficient) and a concept for stabilizing the active phase
outcome. With this procedure[351] neither a structurally pure under the reaction conditions is also necessary. The latter
phase[352] nor a uniform and optimized surface structure will must be achieved, however, without shutting down the
be achieved in general. Errors can arise especially with dynamics of the process (reduction of the surface area is not
combinatorial optimizations in advanced generations[265a, 353] enough). Currently there are only few studies that have been
because the assumption that the catalytic effect is dependent conceived in this way.[117, 209f,h] This opens a field for the
only on the cation composition of the contact mass is development of inorganic chemistry[357] that can simultane-
incorrect. This does not diminish the value of high-throughput ously offer multiple properties of a material arising out of its
procedures for, as an example, the efficient investigation of different dimensions.[48, 51l, 358]
the kinetic parameter range of a system[228a] or for material
development in a predetermined compositional space.[211a, 354]
The goal for the future planning of synthesis strategies 14. Conclusions
should be to start from the characteristics of the desired active
centers. We have learned here that there are essentially two This survey of the current state of heterogeneous catalysis
methods of approach for this. One is that either static or science shows that with the standard model we indeed possess
dynamic centers can be put to use. In catalytic reactions where a firm and generalized conceptual basis. The development of
a component that is difficult to activate (O2, H2, N2, CO2, a quantum mechanical treatment of the atomic aspects of
H2O) is to be converted in dissociated form, using a reactive catalysis based on accurate calculations of the electronic
structure, into a thermodynamically unfavorable product, it is structure of surfaces was triggered by experiments that were
conceivable, as a synthetic approach, to view the combination qualitatively instructional and quantitatively rigorously car-
of both centers as a form of “bifunctionality”. To support the ried out. This treatment, with regards to the state of develop-
concept of such systems that are regularly represented in ment, is at the same stage as the physiochemical theory of
practice, but often unrecognized in their bifunctionality, the catalysis. A slight hesitation accompanies the use of the word
features of both types of centers are given in Table 6. It is “theory” because an important attribute of this level of
evident from this that a significant importance is to be given maturity of a scientific field, the ability to predict phenomena
to the kinetics of the individual production steps[109a, 210a] of the for the concrete knowledge-based development of catalysts
cataylst. In particular, control of the chemical potential of the and processes, has not yet been established.
environment is critical during the steps of thermal treatment The author apologizies to the readers for the often
such as drying, calcination, and annealing. However, these are intentional and unintentional simplification of the presenta-
often not sufficiently analyzed and controlled, at least in tion. The task in this Review was to demonstrate to the reader
academic synthesis. the need to allow a degree of chemical complexity to enter
A planned synthesis, therefore, needs the basic composi- into the rigorous world of physicochemical catalysis. Consid-
tion of a matrix phase that must be produced homogeneously ering catalysis as a rigid gas–solid interfacial problem is only
and as a pure phase. In the case of complex compounds, true for the boundary case of minimal reactivity. Synthetic
however, difficult challenges await[355] to arrive at such phases. chemistry is more interested in developing catalysis into
This does not only hold for bulk phase catalysts but also for a state of predicting and designing functional material of
supported systems in which the precursor of the active phase considerable complexity for performing effectively the
should be brought onto the supporting phase as homoge- desired task of a chemical transformation. This work should
neously as possible. These challenges are solvable, as exem- show that bringing together these two views is a central
plified by the synthesis of micro- and mesoporous system- paradigm in “unifying concepts in catalysis”.

Table 6: Several properties of active centers.

Property Static center Dynamic center
sormation during synthesis in contact with reactants
activation not necessary matrix during initial activation, active sites contnuously
structure static, observable dynamical, “amorphous”
composition similar to bulk phase no relation to bulk phase
existence always as high-energy site stochastic as a rare event
effect splitting of small molecules formation of sensitive molecules
regeneration through desorption through molecular dynamics
component of the bulk phase a termination layer different from the bulk phase (matrix)
realization steps, lattice defects, doping two-dimensional layer or islands, semiconductor properties defined through under layer

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3505
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

In the last few decades, three lines of development in 2. For this reason we must provide materials with the
catalysis science have led to the expectation that we are on the possibility and ability to form dynamic phases. Nano-
cusp of moving from a concept of catalysis to a theory of structuring, chemical and morphological homogeneity,
catalysis. One of these lines is the progressing insight that and careful optimization of the activation are to be
there are no fundamental differences between the subdisci- undertaken for every system and then analytically under-
plines of catalysis science. The treatment of heterogeneous stood.
catalysts within the concept of dynamic systems marked the 3. Catalysts are to be thoroughly characterized before and
first time that the original difference between dynamic active after the reaction.
centers in molecular catalysis and static rigidly defined 4. In situ investigations with complementary methods are
centers in interface analysis fell away. A second line is the not a “luxury” but rather essential parts of the formulat-
swift development of model catalysis away from such static ing a hypothesis on the function of materials.
systems toward nanostructured systems. Here the chemical 5. The development and optimization of the material and
complexity and the dynamic behavior of high-performance reactor as conceptual unit is a prerequisite for the
catalysts can be described without losing the functional understanding of functionality. The indispensable basis
analysis with exactly defined boundary conditions. The third for this is the view of that the investigated transformation
line can be described through the continuing maturity of is a network of reactions both between reactants
in situ functional analysis (or operando analysis). Today we themselves and between the reactants and the catalyst.
are able to decode the function of high-performance catalysts 6. The chemistry that, starting with solid bodies, makes
without the use of models and can investigate the structural dynamic systems out of a stabilizing matrix and an active
features of these functions. interface is still relatively uncharted. Fundamental
These three developments, which have been described knowledge and insight are missing from inorganic
and discussed in the current Review, are focused on ascribing chemistry describing the reactivity of nanostructured
dynamic properties to catalytic materials. The dynamics systems under the conditions of catalytic transformation.
include charge carriers in the electronic structure and atoms 7. Active centers are rare objects in the space–time devel-
in the geometric structure on differing scales in space and opment of a catalyst and only very small stationary
time and can, therefore, be understood and quantified concentrations of them can be expected. The strategy for
through a single characteristic. The task of catalysis to convert investigating them must take this into account. We need
measurable quantities of material is a property that plays out methods that can unambiguously differentiate tiny
on different scales and necessitates for this reason the regions at an interface from the main phase or from an
observation of a large number of physical and chemical under layer.
phenomena simultaneously. 8. Such in situ methods are still scarce. A significant
Viewed historically, we have mastered the challenges of expansion of our analytical palette is still required that
catalysis as system chemistry and multiscale phenomenon by can combine atomic resolution with exact chemical
implementing a series of strict boundary conditions and identification, surface sensitivity, and in situ capabilities.
simplifications in the analysis of catalytic systems excluding 9. Testing materials requires an individual optimization of
complexity. These enable the formulation and justification for the reaction conditions; the practice of testing entire
the standard model. However, to complete the development libraries of materials with universal conditions without
of the theory of catalysis we must take back several of these parameter variation causes important insight to be lost in
boundary conditions and be willing to accept the resulting the reaction procedure and hampers the identification of
heightened complexity of the treatment. This is achieved for possibly new and suitable candidate materials.
example by explicitly implementing the dynamics discussed 10. Concepts for synthesis may be developed along the lines
here and integrating them quantitatively into the mathemat- that materials should be made available in scalable
ical description of catalysis as well as qualitatively into the processes showing a stationary catalytic performance so
design methods. This should not give the impression that this that they may be kinetically and functionally investi-
projection is a new or even an original assertion of this gated. Nonscalable systems and mere curious properties
Review. More or less this same requirement is found in many are less suitable to help catalysis science progress.
conceptional works on catalysis. In the past it was not
conceivable to deal with the complexity of dynamic catalysts The question may be asked why everything seems to have
and at present it is still inconvenient to do so as the efforts become so complicated in the end and which simple force is
required are still substantial. driving all of this complexity. The qualitative answer to this is
At this point the challenge can best be overcome on the simple: the driving force is the minimization of the total
theoretical side if the two orthogonally running approaches to energy of the system. Catalysts are frustrated because we
theoretical treatments of catalysts were to gather into a single ensure during synthesis that they cannot reach their minimum
all-encompassing model instead of remaining opposed to one energy state as long as they are subject to reaction conditions.
another. In experiments we should mind the following points Through this we force a state of the material that is not in
as a consequence of the insight into the dynamic nature of equilibrium under working conditions. The materials attempt
catalysis: to reach equilibrium sets the dynamics in motion: The
1. Active centers cannot be premade but are formed during reaction conditions allow the heterogeneous system on the
the activation of catalyst. surface to begin a process that leads to the thermodynamically

3506 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

stable (deactivated) states via the formation of the active

phase. We can cleverly stop this process from causing changes [1] a) L. M. Aparicio, J. A. Dumesic, Top. Catal. 1994, 1, 233 – 252;
b) L. M. Aparicio, S. A. Rossini, D. G. Sanfilippo, J. E.
in the termination layer or the nanostructuring that go beyond
Rekoske, A. A. Trevino, J. A. Dumesic, Ind. Eng. Chem. Res.
simply returning to the initial phase. The chemical dynamics 1991, 30, 2114 – 2123.
of the formation of active centers is for the material an [2] a) V. Prasad, A. M. Karim, A. Arya, D. G. Vlachos, Ind. Eng.
unwanted consequence of its high-energy state as are the Chem. Res. 2009, 48, 5255 – 5265; b) J. Sun, J. W. Thybaut, G. B.
occasional macroscopic kinetic oscillations in the failed Marin, Catal. Today 2008, 137, 90 – 102; c) B. Frank, F. C.
attempt of the catalyst to stabilize itself. The cyclic process Jentoft, H. Soerijanto, J. Krçhnert, R. Schlçgl, R. Schomcker,
of formation and dissolution of ligand-to-metal bonds in J. Catal. 2007, 246, 177 – 192; d) A. Schle, O. Shekhah, W.
Ranke, R. Schlçgl, G. Kolios, J. Catal. 2005, 231, 173 – 180; e) J.
metal-organic catalysis is also an expression of the energeti-
Libuda, H. J. Freund, Surf. Sci. Rep. 2005, 57, 157 – 298; f) P.
cally unstable overall state of a reaction mixture allowing the Stoltze, Prog. Surf. Sci. 2000, 65, 65 – 150; g) S. Dahl, J.
interchange of different configurations of ligands, reactants, Sehested, C. J. H. Jacobsen, E. Tçrnqvist, I. Chorkendorff, J.
and central atoms. The entire central task of catalysis is, Catal. 2000, 192, 391 – 399; h) C. T. Cambell, H. Topsøe, M.
therefore, the generation and stabilization of energetically Boudart, J. K. Nørskov, Micro- and Macro-Kinetics: Their
frustrated systems. The heat of reaction continuously adds Relationship in Heterogeneous Catalysis, J. C. Baltzer AG, 1994;
new energy to the system and allows the frustrated state to i) O. Deutschmann, L. D. Schmidt, AICHE J. 1998, 44, 2465 –
remain, as long as we do not go beyond a critical threshold 2477.
[3] a) T. Bligaard, J. K. Nørskov, S. Dahl, J. Matthiesen, C. H.
that will send the system into its stable state that is non-
Christensen, J. Sehested, J. Catal. 2004, 224, 206 – 217; b) J.
reactive. Greeley, M. Mavrikakis, J. Am. Chem. Soc. 2004, 126, 3910 –
The collaborative and interdisciplinary way in which 3919; c) J. Libuda, H. J. Freund, J. Phys. Chem. B 2002, 106,
catalysis is currently developing, a trend reflected in the 4901 – 4915.
establishment of successful centers for catalysis research, [4] a) J. Sehested, C. J. H. Jacobsen, E. Tçrnqvist, S. Rokni, P.
brings us closer to the goal of working in a knowledge-based Stoltze, J. Catal. 1999, 188, 83 – 89; b) T. Engel, G. Ertl, J. Chem.
approach. Each one of us can help with the advancement Phys. 1978, 69, 1267 – 1281.
[5] a) J. Thomas, R. Raja, Top. Catal. 2006, 40, 3 – 17; b) J. M.
toward a theory-based science. This happens in our daily work
Thomas, R. Schlçgl, Angew. Chem. Int. Ed. Engl. 1994, 33, 308 –
and through the publication of high-quality studies, although 311; Angew. Chem. 1994, 106, 316 – 319; c) S. L. Wegener, T. J.
two improvements may be suggested. Our colleagues engaged Marks, P. C. Stair, Acc. Chem. Res. 2012, 45, 206 – 214; d) K.
in theoretical work could calibrate their methods with Ariga, A. Vinu, Y. Yamauchi, Q. Ji, J. P. Hill, Bull. Chem. Soc.
generally accepted test systems with documentation for Jpn. 2012, 85, 1 – 32; e) E. L. Lee, I. E. Wachs, J. Phys. Chem. C
every case. They may also establish a link to experimental 2008, 112, 20418 – 20428; f) D. Vitry, J. L. Dubois, W. Ueda, J.
work by calculating observable quantities with their models, Mol. Catal. A 2004, 220, 67 – 76.
again with helpful documentation. On the experimental side [6] S. Schauermann, N. Nilius, S. Shaikhutdinov, H. J. Freund, Acc.
Chem. Res. 2013, 46, 1673 – 1681.
we should begin to document the test results of our catalysts
[7] a) D. Rçthlisberger, O. Khersonsky, A. M. Wollacott, L. Jiang,
in a complete way that can be re-used when needed. A “best J. DeChancie, J. Betker, J. L. Gallaher, E. A. Althoff, A.
practice list” can serve us well for this and should be Zanghellini, O. Dym, S. Albeck, K. N. Houk, D. S. Tawfik, D.
developed from a central body representing our scientific Baker, Nature 2008, 453, 190 – U194; b) A. P. Wight, M. E.
field. If we can cooperatively create structured documenta- Davis, Chem. Rev. 2002, 102, 3589 – 3613; c) F. Besenbacher, I.
tion using the methods of the internet, we will save ourselves Chorkendorff, B. S. Clausen, B. Hammer, A. M. Molenbroek,
much work while at the same time supporting cooperation. J. K. Nørskov, I. Stensgaard, Science 1998, 279, 1913 – 1915.
[8] J. A. Dumesic, A. A. Trevino, J. Catal. 1989, 116, 119 – 129.
Such initiative has proven to be well worth the effort in other
[9] a) J. K. Nørskov, M. Scheffler, H. Toulhoat, MRS Bull. 2006, 31,
fields of science. 669 – 674; b) C. Stampfl, H. J. Kreuzer, S. H. Payne, H. Pfnur,
In the end, the observation remains that we have indeed M. Scheffler, Phys. Rev. Lett. 1999, 83, 2993 – 2996; c) R. A.
come a long way in fulfilling the demands introduced in the Vansanten, M. Neurock, Catal. Rev. Sci. Eng. 1995, 37, 557 –
two citations at the beginning of this work. The importance of 698.
catalysis as a cross-disciplinary science of chemistry for our [10] a) D. Mei, P. A. Sheth, M. Neurock, C. M. Smith, J. Catal. 2006,
discipline and for technological challenges is clear. It should 242, 1 – 15; b) M. Neurock, S. A. Wasileski, D. Mei, Chem. Eng.
Sci. 2004, 59, 4703 – 4714; c) K. Reuter, M. Scheffler, Phys. Rev.
be incentive enough to complete our work and to proceed
B 2006, 73, 045433.
into the future with a physics-based comprehensive and
[11] a) K. Reuter, M. Scheffler, Phys. Rev. Lett. 2003, 90, 046103;
predictive theory of catalysis that can guide rational synthesis b) C. Stampfl, M. V. Ganduglia-Pirovano, K. Reuter, M.
to arrive at suitable catalysis solutions for the challenges in Scheffler, Surf. Sci. 2002, 500, 368 – 394.
our future. [12] K. Kwapien, M. Sierka, J. Doebler, J. Sauer, ChemCatChem
2010, 2, 819 – 826.
[13] a) V. Schott, H. Oberhofer, A. Birkner, M. Xu, Y. Wang, M.
Acknowlegments Muhler, K. Reuter, C. Wçll, Angew. Chem. Int. Ed. 2013, 52,
11925 – 11929; Angew. Chem. 2013, 125, 12143 – 12147; b) S.
Matera, K. Reuter, J. Catal. 2012, 295, 261 – 268; c) X.
Many thanks to my present and former co-workers. Rozanska, J. Sauer, J. Phys. Chem. A 2009, 113, 11586 – 11594.
[14] a) W. Piskorz, F. Zasada, P. Stelmachowski, A. Kotarba, Z.
Received: November 4, 2014 Sojka, J. Phys. Chem. C 2013, 117, 18488 – 18501; b) M. Saeys,
Published online: February 18, 2015 M. F. Reyniers, M. Neurock, G. B. Marin, J. Phys. Chem. B

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3507
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

2005, 109, 2064 – 2073; c) M. Neurock, W. T. Tysoe, Top. Catal. [27] a) M. G. Willinger, W. Zhang, O. Bondarchuk, S. Shaikhutdi-
2013, 56, 1314 – 1332; d) Y. H. Chin, C. Buda, M. Neurock, E. nov, H. J. Freund, R. Schlçgl, Angew. Chem. Int. Ed. 2014, 53,
Iglesia, J. Am. Chem. Soc. 2011, 133, 15958 – 15978; e) I. C. 5998 – 6001; Angew. Chem. 2014, 126, 6108 – 6112; b) S. Kuld, C.
Man, H.-Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I. Martinez, Conradsen, P. G. Moses, I. Chorkendorff, J. Sehested, Angew.
N. G. Inoglu, J. Kitchin, T. F. Jaramillo, J. K. Nørskov, J. Chem. Int. Ed. 2014, 53, 5941 – 5945; Angew. Chem. 2014, 126,
Rossmeisl, ChemCatChem 2011, 3, 1159 – 1165. 6051 – 6055; c) H. J. Freund, N. Nilius, T. Risse, S. Schauer-
[15] a) R. Horn, O. Korup, M. Geske, U. Zavyalova, I. Oprea, R. mann, Phys. Chem. Chem. Phys. 2014, 16, 8148 – 8167; d) V.
Schlçgl, Rev. Sci. Instrum. 2010, 81, 064102; b) Y. Simon, F. Papaefthimiou, T. Dintzer, V. Dupuis, A. Tamion, F. Tournus,
Baronnet, G. M. Come, P. M. Marquaire, Nat. Gas Convers. VII A. Hillion, D. Teschner, M. Havecker, A. Knop-Gericke, R.
2004, 147, 571 – 576; c) M. Baerns, O. V. Buyevskaya, Erdçl Schlçgl, S. Zafeiratos, ACS Nano 2011, 5, 2182 – 2190; e) J.
Erdgas Kohle 2000, 116, 25 – 30; d) O. Deutschmann, F. Strunk, W. C. Vining, A. T. Bell, J. Phys. Chem. C 2010, 114,
Behrendt, J. Warnatz, Catal. Today 1994, 21, 461 – 470. 16937; f) P. Gonzlez-Navarrete, L. Gracia, M. Calatayud, J.
[16] P. Stoltze, J. K. Nørskov, Phys. Rev. Lett. 1985, 55, 2502 – 2505. Andrs, J. Phys. Chem. C 2010, 114, 6039 – 6046; g) P. Maity,
[17] C. T. Campbell, Top. Catal. 1994, 1, 353 – 366. C. S. Gopinath, S. Bhaduri, G. K. Lahiri, Green Chem. 2009, 11,
[18] J. Vill, A. Warshel, J. Phys. Chem. B 2001, 105, 7887 – 7907. 554 – 561; h) H.-S. Kim, S. A. Zygmunt, P. C. Stair, P. Zapol,
[19] D. E. Koshland, Proc. Natl. Acad. Sci. USA 1958, 44, 98 – 104. L. A. Curtiss, J. Phys. Chem. C 2009, 113, 8836 – 8843; i) P. F.
[20] a) V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283 – Fulvio, S. Pikus, M. Jaroniec, ACS Appl. Mater. Interfaces 2010,
315; b) S. J. Connon, S. Blechert, Angew. Chem. Int. Ed. 2003, 2, 134 – 142; j) G. L. Bezemer, J. H. Bitter, H. Kuipers, H.
42, 1900 – 1923; Angew. Chem. 2003, 115, 1944 – 1968; c) W. A. Oosterbeek, J. E. Holewijn, X. D. Xu, F. Kapteijn, A. J.
Herrmann, Angew. Chem. Int. Ed. 2002, 41, 1290 – 1309; van Dillen, K. P. de Jong, J. Am. Chem. Soc. 2006, 128, 3956 –
Angew. Chem. 2002, 114, 1342 – 1363; d) R. Noyori, T. 3964; k) M. Ruitenbeek, A. J. van Dillen, F. M. F. de Groot,
Ohkuma, Angew. Chem. Int. Ed. 2001, 40, 40 – 73; Angew. I. E. Wachs, J. W. Geus, D. C. Koningsberger, Top. Catal. 2000,
Chem. 2001, 113, 40 – 75; e) B. M. Trost, Angew. Chem. Int. Ed. 10, 241 – 254.
Engl. 1995, 34, 259 – 281; Angew. Chem. 1995, 107, 285 – 307; [28] a) S. Lambert, N. Job, L. DSouza, M. F. Ribeiro Pereira, R.
f) C. A. Tolman, Chem. Rev. 1977, 77, 313 – 348. Pirard, B. Heinrichs, J. L. Figueiredo, J.-P. Pirard, J. R. Regal-
[21] a) K. Topsøe, C. V. Ovesen, B. S. Clausen, N. Y. Topsøe, P. E. H. buto, J. Catal. 2009, 261, 23 – 33; b) W. A. Spieker, J. Liu, J. T.
Nielsen, E. Tçrnqvist, J. K. Norskov in Dynamics of Surfaces Miller, A. J. Kropf, J. R. Regalbuto, Appl. Catal. A 2002, 232,
and Reaction Kinetics in Heterogeneous Catalysis, Vol. 109, 219 – 235; c) J. R. Regalbuto, A. Navada, S. Shadid, M. L.
1997, S. 121 – 139; b) T. C. R. Rocha, A. Oestereich, D. V. Bricker, Q. Chen, J. Catal. 1999, 184, 335 – 348; d) K. B. Agashe,
Demidov, M. Hvecker, S. Zafeiratos, G. Weinberg, V. I. J. R. Regalbuto, J. Colloid Interface Sci. 1997, 185, 174 – 189;
Bukhtiyarov, A. Knop-Gericke, R. Schlçgl, Phys. Chem. e) J. R. Regalbuto, T. H. Fleisch, E. E. Wolf, J. Catal. 1987, 107,
Chem. Phys. 2012, 14, 4554 – 4564; c) A. Haghofer, K. Fot- 114 – 128.
tinger, F. Girgsdies, D. Teschner, A. Knop-Gericke, R. Schlçgl, [29] a) K. Amakawa, L. L. Sun, C. S. Guo, M. Hvecker, P. Kube,
G. Rupprechter, J. Catal. 2012, 286, 13 – 21; d) Z. Feng, L. I. E. Wachs, S. Lwin, A. I. Frenkel, A. Patlolla, K. Hermann, R.
Cheng, C.-Y. Kim, J. W. Elam, Z. Zhang, L. A. Curtiss, P. Zapol, Schlçgl, A. Trunschke, Angew. Chem. Int. Ed. 2013, 52, 13553 –
M. J. Bedzyk, J. Phys. Chem. Lett. 2012, 3, 2845 – 2850; e) G. 13557; Angew. Chem. 2013, 125, 13796 – 13800; b) P. Gruene, T.
Ertl, Science 1991, 254, 1750 – 1755; f) F. Cavani, G. Centi, F. Wolfram, K. Pelzer, R. Schlçgl, A. Trunschke, Catal. Today
Trifiro, R. K. Grasselli, Catal. Today 1988, 3, 185 – 198; g) H. 2010, 157, 137 – 142; c) R. Lamber, N. I. Jaeger, A. Trunschke,
Nair, J. T. Miller, E. A. Stach, C. D. Baertsch, J. Catal. 2010, H. Miessner, Catal. Lett. 1991, 11, 1 – 9.
270, 40 – 47; h) O. Ovsitser, M. Cherian, A. Brckner, E. V. [30] a) A. K. Datye, Q. Xu, K. C. Kharas, J. M. McCarty, Catal.
Kondratenko, J. Catal. 2009, 265, 8 – 18; i) R. Imbihl, Surf. Sci. Today 2006, 111, 59 – 67; b) A. K. Datye, J. Bravo, T. R. Nelson,
2009, 603, 1671 – 1679; j) V. Salehoun, A. Khodadadi, Y. P. Atanasova, M. Lyubovsky, L. Pfefferle, Appl. Catal. A 2000,
Mortazavi, A. Talebizadeh, Chem. Eng. Sci. 2008, 63, 4910 – 198, 179 – 196; c) A. K. Datye, D. S. Kalakkad, M. H. Yao, D. J.
4916. Smith, J. Catal. 1995, 155, 148 – 153; d) H. J. Freund, M. Heyde,
[22] a) D. Rosenthal, F. Girgsdies, O. Timpe, G. Weinberg, R. N. Nilius, S. Schauermann, S. Shaikhutdinov, M. Sterrer, J.
Schlçgl, Z. Phys. Chem.-Int. J. Res. Phys. Chem. Chem. Phys. Catal. 2013, 308, 154 – 167.
2011, 225, 57 – 68; b) I. Bçttger, B. Pettinger, T. Schedel- [31] a) T. Inasaki, S. Kobayashi, Electrochim. Acta 2009, 54, 4893 –
Niedrig, A. Knop-Gericke, R. Schlçgl, G. F. Froment, G. C. 4897; b) A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. Int.
Waugh, Self Sustained Oscillations Over Copper in the Catalytic Ed. 2006, 45, 7896 – 7936; Angew. Chem. 2006, 118, 8064 – 8105;
Oxidation of Methanol, Elsevier, Amsterdam, 2001; c) H. c) L. R. Radovic, C. Moreno-Castilla, J. Rivera-Utrilla, Chem.
Werner, D. Herein, G. Schulz, U. Wild, R. Schlçgl, Catal. Phys. Carbon 2001, 27, 227 – 405; d) C. Bianchi, F. Porta, L.
Lett. 1997, 49, 109 – 119; d) G. Ertl, Angew. Chem. Int. Ed. 2008, Prati, M. Rossi, Top. Catal. 2000, 13, 231 – 236; e) M. Haruta,
47, 3524 – 3535; Angew. Chem. 2008, 120, 3578 – 3590. Catal. Today 1997, 36, 153 – 166.
[23] a) R. K. Grasselli, C. G. Lugmair, A. F. Volpe, Jr., Top. Catal. [32] a) N. Mizuno, M. Misono, Chem. Rev. 1998, 98, 199 – 218; b) J.-
2008, 50, 66 – 73; b) W. Ueda, Y. Endo, N. Watanabe, Top. Catal. P. Tessonnier, S. Goubert-Renaudin, S. Alia, Y. Yan, M. A.
2006, 38, 261 – 268; c) G. H. Dai, Q. J. Yan, Y. Wang, Q. S. Liu, Barteau, Langmuir 2013, 29, 393 – 402.
Chem. Phys. 1991, 155, 275 – 284; d) E. W. McFarland, H. [33] a) H. B. Buergi, J. D. Dunitz, Acc. Chem. Res. 1983, 16, 153 –
Metiu, Chem. Rev. 2013, 113, 4391 – 4427. 161; b) S. R. Leone, M. Ahmed, K. R. Wilson, Phys. Chem.
[24] a) G. A. Somorjai, J. Y. Park, Surf. Sci. 2009, 603, 1293 – 1300; Chem. Phys. 2010, 12, 6564 – 6578.
b) F. Zaera, Prog. Surf. Sci. 2001, 69, 1 – 98. [34] a) J. C. Tully, Annu. Rev. Phys. Chem. 2000, 51, 153 – 178;
[25] a) M. Baron, O. Bondarchuk, D. Stacchiola, S. Shaikhutdinov, b) C. T. Rettner, D. J. Auerbach, J. C. Tully, A. W. Kleyn, J.
H. J. Freund, J. Phys. Chem. C 2009, 113, 6042 – 6049; b) M. Phys. Chem. 1996, 100, 13021 – 13033.
Chen, C. M. Friend, E. Kaxiras, J. Am. Chem. Soc. 2001, 123, [35] a) I. R. Epstein, K. Showalter, J. Phys. Chem. 1996, 100, 13132 –
2224 – 2230. 13147; b) J. M. Edmond, A. Spivack, B. C. Grant, M. H. Hu,
[26] X. Y. Sun, B. Li, H. Metiu, J. Phys. Chem. C 2013, 117, 7114 – Z. X. Chen, S. Chen, X. S. Zeng, Cont. Shelf Res. 1985, 4, 17 –
7122. 36; c) D. Imre, J. L. Kinsey, A. Sinha, J. Krenos, J. Phys. Chem.

3508 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

1984, 88, 3956 – 3964; d) A. C. Luntz, J. Chem. Phys. 1980, 73, Gibson, W. A. Goddard, D. W. Goodman, J. Keller, G. J. Kubas,
1143 – 1152. H. H. Kung, J. E. Lyons, L. E. Manzer, T. J. Marks, K.
[36] a) B. M. Weckhuysen, Phys. Chem. Chem. Phys. 2003, 5, 4351 – Morokuma, K. M. Nicholas, R. Periana, L. Que, J. Rostrup-
4360; b) D. Mores, E. Stavitski, M. H. F. Kox, J. Kornatowski, Nielson, W. M. H. Sachtler, L. D. Schmidt, A. Sen, G. A.
U. Olsbye, B. M. Weckhuysen, Chem. Eur. J. 2008, 14, 11320 – Somorjai, P. C. Stair, B. R. Stults, W. Tumas, Chem. Rev. 2001,
11327; c) C. Mattevi, C. T. Wirth, S. Hofmann, R. Blume, M. 101, 953 – 996; b) J. M. Thomas, Energy Environ. Sci. 2014, 7,
Cantoro, C. Ducati, C. Cepek, A. Knop-Gericke, S. Milne, C. 19 – 20; c) N. S. Lewis, D. G. Nocera, Proc. Natl. Acad. Sci. USA
Castellarin-Cudia, S. Dolafi, A. Goldoni, R. Schlçgl, J. Rob- 2006, 103, 15729 – 15735; d) G. Centi, S. Perathoner, Catal.
ertson, J. Phys. Chem. C 2008, 112, 12207 – 12213; d) D. Today 2009, 148, 191 – 205; e) R. Schlçgl, Angew. Chem. Int.
Teschner, E. Vass, M. Hvecker, S. Zafeiratos, P. Schnçrch, H. Ed. 2011, 50, 6424 – 6426; Angew. Chem. 2011, 123, 6550 – 6553.
Sauer, A. Knop-Gericke, R. Schlçgl, M. Chamam, A. Wootsch, [43] a) N. Cox, D. A. Pantazis, F. Neese, W. Lubitz, Acc. Chem. Res.
A. S. Canning, J. J. Gamman, S. D. Jackson, J. McGregor, L. F. 2013, 46, 1588 – 1596; b) W. Lubitz, E. J. Reijerse, J. Messinger,
Gladden, J. Catal. 2006, 242, 26 – 37; e) O. Shekhah, W. Ranke, Energy Environ. Sci. 2008, 1, 15 – 31.
A. Schle, G. Kolios, R. Schlçgl, Angew. Chem. Int. Ed. 2003, [44] a) J. Rossmeisl, Z. W. Qu, H. Zhu, G. J. Kroes, J. K. Norskov, J.
42, 5760 – 5763; Angew. Chem. 2003, 115, 5938 – 5941; f) T. Electroanal. Chem. 2007, 607, 83 – 89; b) J. Rossmeisl, K.
Schedel-Niedrig, M. Hvecker, G. Knop, R. Schlçgl, Phys. Dimitrievski, P. Siegbahn, J. K. Norskov, J. Phys. Chem. C
Chem. Chem. Phys. 2000, 2, 3473 – 3481; g) Y. Yang, Z. G. Lin, 2007, 111, 18821 – 18823.
J. Electroanal. Chem. 1994, 364, 23 – 30. [45] a) T.-P. Fellinger, F. Hasche, P. Strasser, M. Antonietti, J. Am.
[37] a) D. S. Su, S. Perathoner, G. Centi, Chem. Rev. 2013, 113, Chem. Soc. 2012, 134, 4072 – 4075; b) R. Arrigo, M. E. Schuster,
5782 – 5816; b) O. Khavryuchenko, B. Frank, A. Trunschke, K. S. Abate, S. Wrabetz, K. Amakawa, D. Teschner, M. Freni, G.
Hermann, R. Schlçgl, J. Phys. Chem. C 2013, 117, 6225 – 6234; Centi, S. Perathoner, M. Hvecker, R. Schlçgl, ChemSusChem
c) J. J. Delgado, X. Chen, B. Frank, D. S. Su, R. Schlçgl, Catal. 2014, 7, 179 – 194; c) S. Abate, R. Arrigo, M. E. Schuster, S.
Today 2012, 186, 93 – 98; d) X. Liu, B. Frank, W. Zhang , T. P. Perathoner, G. Centi, A. Villa, D. S. Su, R. Schlçgl, Catal.
Cotter, R. Schlçgl, D. S. Su, Angew. Chem. Int. Ed. 2011, 50, Today 2010, 157, 280 – 285.
3318 – 3322; Angew. Chem. 2011, 123, 3376 – 3380; e) B. Frank, [46] H.-J. Freund, G. Meijer, M. Scheffler, R. Schlçgl, M. Wolf,
R. Blume, A. Rinaldi, A. Trunschke, R. Schlçgl, Angew. Chem. Angew. Chem. Int. Ed. 2011, 50, 10064 – 10094; Angew. Chem.
Int. Ed. 2011, 50, 10226 – 10230; Angew. Chem. 2011, 123, 2011, 123, 10242 – 10275.
10408 – 10413; f) D. S. Su, J. Zhang, A. Thomas, X. Wang, J. [47] a) N. Dietl, M. Schlangen, H. Schwarz, Angew. Chem. Int. Ed.
Paraknowitsch, R. Schlçgl, ChemSusChem 2010, 3, 169 – 180; 2012, 51, 5544 – 5555; Angew. Chem. 2012, 124, 5638 – 5650;
g) D. S. Su, R. Schlçgl, ChemSusChem 2010, 3, 136 – 168; h) J. b) K. Takanabe, E. Iglesia, J. Phys. Chem. C 2009, 113, 10131 –
McGregor, Z. Huang, E. P. J. Parrott, J. A. Zeitler, K. L. 10145; c) M. Geske, K. Pelzer, R. Horn, F. C. Jentoft, R.
Nguyen, J. M. Rawson, A. Carley, T. W. Hansen, J.-P. Tessonier, Schlçgl, Catal. Today 2009, 142, 61 – 69.
D. S. Su, D. Teschner, E. M. Vass, A. Knop-Gericke, R. Schlçgl, [48] R. Schlçgl, S. B. Abd Hamid, Angew. Chem. Int. Ed. 2004, 43,
L. F. Gladden, J. Catal. 2010, 269, 329 – 339; i) B. Frank, M. 1628 – 1637; Angew. Chem. 2004, 116, 1656 – 1667.
Morassutto, R. Schomcker, R. Schlçgl, D. S. Su, Chem- [49] a) J. Y. Park, G. A. Somorjai, ChemPhysChem 2006, 7, 1409 –
CatChem 2010, 2, 644 – 648; j) J. Zhang, X. Wang, Q. Su, L. 1413; b) A. Lloyd Spetz, P. Tobias, L. Uneus, H. Svenningstorp,
Zhi, A. Thomas, X. Feng, D. S. Su, R. Schlçgl, K. Mllen, J. Am. L. G. Ekedahl, L. Lundstrom, Sens. Actuators B 2000, 70, 67 –
Chem. Soc. 2009, 131, 11296 – 11297; k) J.-P. Tessonnier, A. 76.
Villa, O. Majoulet, D. S. Su, R. Schlçgl, Angew. Chem. Int. Ed. [50] a) A. Bienholz, R. Blume, A. Knop-Gericke, F. Girgsdies, M.
2009, 48, 6543 – 6546; Angew. Chem. 2009, 121, 6665 – 6668; Behrens, P. Claus, J. Phys. Chem. C 2011, 115, 999 – 1005; b) S.
l) B. Frank, J. Zhang, R. Blume, R. Schlçgl, D. S. Su, Angew. Arndt, U. Simon, S. Heitz, A. Berthold, B. Beck, O. Goerke,
Chem. Int. Ed. 2009, 48, 6913 – 6917; Angew. Chem. 2009, 121, J. D. Epping, T. Otremba, Y. Aksu, E. Irran, G. Laugel, M.
7046 – 7051; m) J. Zhang, X. Liu, A. Blume, R. Schlçgl, D. S. Su, Driess, H. Schubert, R. Schomaecker, Top. Catal. 2011, 54,
Science 2008, 322, 73 – 77; n) A. Thomas, A. Fischer, F. 1266 – 1285; c) Z. Pal, R. Schlçgl, Surf. Sci. 2009, 603, 1793 –
Goettmann, M. Antonietti, J.-O. Mller, R. Schlçgl, J. M. 1801; d) T. Kim, G. Liu, M. Boaro, S. I. Lee, J. M. Vohs, R. J.
Carlsson, J. Mater. Chem. 2008, 18, 4893 – 4908; o) X. Liu, D. S. Gorte, O. H. Al-Madhi, B. O. Dabbousi, J. Power Sources 2006,
Su, R. Schlçgl, Carbon 2008, 46, 544 – 548; p) J. Zhang, D. S. Su, 155, 231 – 238; e) A. Wootsch, U. Wild, R. Schlçgl, Z. Pal,
A. Zhang, D. Wang, R. Schlçgl, C. Hbert, Angew. Chem. Int. Catal. Commun. 2004, 5, 419 – 423; f) A. Trunschke, J. Deutsch,
Ed. 2007, 46, 7319 – 7323; Angew. Chem. 2007, 119, 7460 – 7464. D. Mueller, H. Lieske, V. Quaschning, E. Kemnitz, Catal. Lett.
[38] F. Zaera, A. J. Gellman, G. A. Somorjai, Acc. Chem. Res. 1986, 2002, 83, 271 – 279; g) J. R. Rostrup-Nielsen, J. Sehested in
19, 24 – 31. Catalyst Deactivation 2001, Proceedings, Vol. 139, 2001, S. 1 –
[39] a) M. Elstner, P. Hobza, T. Frauenheim, S. Suhai, E. Kaxiras, J. 12; h) C. H. Bartholomew, Appl. Catal. A 2001, 212, 17 – 60;
Chem. Phys. 2001, 114, 5149 – 5155; b) M. E. van Leeuwen, B. i) R. Farra, M. Eichelbaum, R. Schlçgl, L. Szentmikl si, T.
Smit, Phys. Rev. Lett. 1993, 71, 3991 – 3994; c) P. W. Carter, Schmidt, A. P. Amrute, C. Mondelli, J. Prez-Ram rez, D.
M. D. Ward, J. Am. Chem. Soc. 1993, 115, 11521 – 11535. Teschner, J. Catal. 2013, 297, 119 – 127; j) W. Ranke, A. Schle,
[40] M. Maestri, K. Reuter, Angew. Chem. Int. Ed. 2011, 50, 1194 – O. Shekhah, G. Kolios, U. Nieken, R. Schlçgl, Book of
1197; Angew. Chem. 2011, 123, 1226 – 1229. Extended Abstracts der DECHEMA e. V. 2006, S. 234 – 238;
[41] a) C. Torborg, M. Beller, Adv. Synth. Catal. 2009, 351, 3027 – k) Z. Pal, A. Wootsch, R. Schlçgl, U. Wild, Appl. Catal. A
3043; b) H. U. Blaser, A. Indolese, F. Naud, U. Nettekoven, A. 2005, 282, 135 – 145; l) B. S. Klose, R. E. Jentoft, T. Ressler, P.
Schnyder, Adv. Synth. Catal. 2004, 346, 1583 – 1598; c) G. W. Joshi, A. Trunschke, R. Schlçgl, F. C. Jentoft, Proc. DGMK Int.
Coates, J. Chem. Soc. Dalton Trans. 2002, 467 – 475; d) J. H. Conf. 2005, 3, 23 – 30; m) N. M. Rodriguez, P. E. Anderson, A.
Clark, Green Chem. 1999, 1, 1 – 8; e) R. A. Sheldon, Chem. Wootsch, U. Wild, R. Schlçgl, Z. Pal, J. Catal. 2001, 197, 365 –
Commun. 2008, 3352 – 3365; f) G. Centi, M. Misono, Catal. 377.
Today 1998, 41, 287 – 296. [51] a) B. M. Weckhuysen, Angew. Chem. Int. Ed. 2009, 48, 4910 –
[42] a) H. Arakawa, M. Aresta, J. N. Armor, M. A. Barteau, E. J. 4943; Angew. Chem. 2009, 121, 5008 – 5043; b) S. J. Tinnemans,
Beckman, A. T. Bell, J. E. Bercaw, C. Creutz, E. Dinjus, D. A. J. G. Mesu, K. Kervinen, T. Visser, T. A. Nijhuis, A. M. Beale,
Dixon, K. Domen, D. L. DuBois, J. Eckert, E. Fujita, D. H. D. E. Keller, A. M. J. van der Eerden, B. M. Weckhuysen,

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3509
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Catal. Today 2006, 113, 3 – 15; c) B. M. Weckhuysen, D. E. e) S. R. Bare, S. D. Kelly, B. Ravel, N. Greenlay, L. King, G. E.
Keller, Catal. Today 2003, 78, 25 – 46; d) A. Brckner, Chem. Mickelson, Phys. Chem. Chem. Phys. 2010, 12, 7702 – 7711.
Commun. 2005, 1761 – 1763; e) T. Kandemir, M. E. Schuster, A. [59] a) T. Ushikubo, Catal. Today 2003, 78, 79 – 84; b) T. Ushikubo,
Senyshyn, M. Behrens, R. Schlçgl, Angew. Chem. Int. Ed. 2013, Catalysts in Petroleum Refining and Petrochemicals, Proceed-
52, 12723 – 12726; Angew. Chem. 2013, 125, 12955 – 12959; f) R. ings [of the] Annual Saudi-Japanese Symposium, 11th, Dhah-
Arrigo, M. Hvecker, M. E. Schuster, C. Ranjan, E. Stotz, A. ran, Saudi Arabia, Nov. 11 – 12, 2001, paper 14/11-paper 14/19;
Knop-Gericke, R. Schlçgl, Angew. Chem. Int. Ed. 2013, 52, c) R. K. Grasselli, J. D. Burrington in Handbook of Heteroge-
11660 – 11664; Angew. Chem. 2013, 125, 11874 – 11879; g) R. neous Catalysis, Vol. 7, 2nd ed. (Eds.: G. Ertl, H. Knçzinger,
Weatherup, B. C. Bayer, R. Blume, C. Ducati, C. Bhtz, R. Jens Weitkamp), Wiley-VCH, Weinheim, 2008, pp. 3479 – 3489.
Schlçgl, S. Hofmann, Nano Lett. 2011, 11, 4154 – 4160; h) R. [60] a) J. B. Wagner, O. Timpe, F. A. Hamid, A. Trunschke, D. S. Su,
Horn, O. Korup, M. Geske, U. Zavyalova, I. Oprea, R. Schlçgl, R. K. Widi, S. B. Abd Hamid, R. Schlçgl, Top. Catal. 2006, 38,
Rev. Sci. Instrum. 2010, 81, 064102; i) M. Eichelbaum, R. 51 – 58; b) W. Zhang, A. Trunschke, R. Schlçgl, D. S. Su, Angew.
Glaum, M. Hvecker, K. Wittich, C. Heine, H. Schwarz, C.-K. Chem. Int. Ed. 2010, 49, 6084 – 6089; Angew. Chem. 2010, 122,
Dobner, C. Welker-Niewoudt, A. Trunschke, R. Schlçgl, 6220 – 6225.
ChemCatChem 2013, 5, 2318 – 2329; j) R. Arrigo, M. Hvecker, [61] M. Hvecker, S. Wrabetz, J. Krçhnert, R. Naumann dAlnon-
M. E. Schuster, C. Ranjan, E. Stotz, A. Knop-Gericke, R. court, Y. V. Kolenko, F. Girgsdies, R. Schlçgl, A. Trunschke, J.
Schlçgl, Angew. Chem. Int. Ed. 2013, 52, 11660 – 11664; Angew. Catal. 2012, 285, 48 – 60.
Chem. 2013, 125, 11874 – 11879; k) M. Behrens, R. Schlçgl in [62] a) N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv. Synth.
Characterization of Solid Materials and Heterogeneous Cata- Catal. 2006, 348, 609 – 679; b) D. Astruc, F. Lu, J. R. Aranzaes,
lysts: From Structure to Surface Reactivity, Vol. 2 (Eds.: J. C. Angew. Chem. Int. Ed. 2005, 44, 7852 – 7872; Angew. Chem.
Vdrine, M. Che), Wiley-VCH, Weinheim, 2012, pp. 611 – 654; 2005, 117, 8062 – 8083; c) A. Corma, H. Garcia, Chem. Rev.
l) R. Schlçgl, Adv. Catal. 2009, 52, 273 – 338; m) R. Arrigo, M. 2003, 103, 4307 – 4365; d) C. Copret, M. Chabanas, R. P. Saint-
Hvecker, S. Wrabetz, R. Blume, M. Lerch, A. Knop-Gericke,
Arroman, J. M. Basset, Angew. Chem. Int. Ed. 2003, 42, 156 –
R. Schlçgl, D. S. Su, J. Am. Chem. Soc. 2010, 132, 9616 – 9630.
181; Angew. Chem. 2003, 115, 164 – 191; e) F. Fache, E. Schulz,
[52] a) J. M. Thomas, Angew. Chem. Int. Ed. Engl. 1994, 33, 913 –
M. L. Tommasino, M. Lemaire, Chem. Rev. 2000, 100, 2159 –
937; Angew. Chem. 1994, 106, 963 – 989; b) H. G. Karge, W.
2231; f) W. A. Herrmann, C. W. Kohlpaintner, Angew. Chem.
Niessen, H. Bludau, Appl. Catal. A 1996, 146, 339 – 349.
Int. Ed. Engl. 1993, 32, 1524 – 1544; Angew. Chem. 1993, 105,
[53] R. Schlçgl in Basic Principles in Applied Catalysis (Ed.: M.
1588 – 1609; g) T. J. Marks, Acc. Chem. Res. 1992, 25, 57 – 65.
Baerns), Springer, Berlin, 2004, pp. 321 – 360.
[63] http://www.unicat.tu-berlin.de/.
[54] a) B. S. Clausen, S. Morup, H. Topsoe, R. Candia, E. J. Jensen,
[64] a) Y. Wang, H. X. Lin, L. Chen, S. Y. Ding, Z. C. Lei, D. Y. Liu,
A. Baranski, A. Pattek, J. Phys. 1976, 37, 245 – 248; b) F.
X. Y. Cao, H. J. Liang, Y. B. Jiang, Z. Q. Tian, Chem. Soc. Rev.
Jentoft, S. Klokishner, J. Krçhnert, J. Melsheimer, T. Ressler, O.
2014, 43, 399 – 411; b) R. F. Ludlow, S. Otto, Chem. Soc. Rev.
Timpe, J. Wienold, R. Schlçgl, Appl. Catal. A 2003, 256, 291 –
2008, 37, 101 – 108.
317; c) J. F. Haw, W. G. Song, D. M. Marcus, J. B. Nicholas, Acc.
[65] J. H. van Esch, B. L. Feringa, Angew. Chem. Int. Ed. 2000, 39,
Chem. Res. 2003, 36, 317 – 326; d) B. Herzog, M. Wohlers, R.
2263 – 2266; Angew. Chem. 2000, 112, 2351 – 2354.
Schlçgl, Microchim. Acta 1997, 14, 703 – 704; e) B. Herzog, D.
[66] J. A. Labinger, J. Mol. Catal. A 2004, 220, 27 – 35.
Bokern, T. Braun, R. Schlçgl, Mater. Sci. Forum 1994, 166 – 169,
517 – 522; f) X. Bao, B. Pettinger, G. Ertl, R. Schlçgl, Ber. [67] O. Korup, C. F. Goldsmith, G. Weinberg, M. Geske, T.
Bunsen-Ges. 1993, 97, 322 – 325; g) A. Trunschke, H. Ewald, H. Kandemir, R. Schlçgl, R. Horn, J. Catal. 2013, 297, 1 – 16.
Miessner, S. Marengo, S. Martinengo, F. Pinna, L. Zanderighi, J. [68] a) U. Zavyalova, G. Weinberg, W. Frandsen, F. Girgsdies, T.
Mol. Catal. 1992, 74, 365 – 377. Risse, K.-P. Dinse, R. Schlçgl, R. Horn, ChemCatChem 2011, 3,
[55] A. M. Beale, A. M. J. van der Eerden, K. Kervinen, M. A. 1779 – 1788; b) S. Arndt, G. Laugel, S. Levchenko, R. Horn, M.
Newton, B. M. Weckhuysen, Chem. Commun. 2005, 3015 – Baerns, M. Scheffler, R. Schçgl, R. Schomcker, Catal. Rev. Sci.
3017. Eng. 2011, 53, 424 – 514.
[56] a) K. Amakawa, L. Sun, C. Guo, M. Hvecker, P. Kube, I. E. [69] a) U. Simon, O. Goerke, A. Berthold, S. Arndt, R. Scho-
Wachs, S. Lwin, A. I. Frenkel, A. Patlolla, K. Hermann, R. maecker, H. Schubert, Chem. Eng. J. 2011, 168, 1352 – 1359;
Schlçgl, A. Trunschke, Angew. Chem. Int. Ed. 2013, 52, 13553 – b) P. Myrach, N. Nilius, S. V. Levchenko, A. Gonchar, T. Risse,
13557; Angew. Chem. 2013, 125, 13796 – 13800; b) D. Teschner, K. P. Dinse, L. A. Boatner, W. Frandsen, R. Horn, H. J. Freund,
R. Farra, L. Yao, R. Schlçgl, H. Soerijanto, R. Schomaeker, T. R. Schlçgl, M. Scheffler, ChemCatChem 2010, 2, 854 – 862.
Schmidt, L. Szentmikl si, A. Amrute, C. Mondelli, J. Prez- [70] C. Batiot, B. K. Hodnett, Appl. Catal. A 1996, 137, 179 – 191.
Ramirez, G. Novell-Leruth, N. L pez, J. Catal. 2012, 285, 273 – [71] U. Zavyalova, M. Holena, R. Schlçgl, M. Baerns, Chem-
284; c) G. Wowsnick, D. Teschner, I. Kasatkin, F. Girgsdies, M. CatChem 2011, 3, 1935 – 1947.
Armbrster, A. Zhang, Y. Grin, R. Schlçgl, M. Behrens, J. [72] J. H. Lunsford, Angew. Chem. Int. Ed. Engl. 1995, 34, 970 – 980;
Catal. 2014, 309, 209 – 220; d) G. Wowsnick, D. Teschner, M. Angew. Chem. 1995, 107, 1059 – 1070.
Armbrster, I. Kasatkin, F. Girgsdies, Y. Grin, R. Schlçgl, M. [73] S. Arndt, G. Laugel, S. Levchenko, R. Horn, M. Baerns, M.
Behrens, J. Catal. 2014, 309, 221 – 230; e) M. Behrens, F. Studt, Scheffler, R. Schlçgl, R. Schomacker, Catal. Rev. Sci. Eng. 2011,
I. Kasatkin, S. Khl, M. Hvecker, F. Abild-Pedersen, S. 53, 424 – 514.
Zander, F. Girgsdies, P. Kurr, B.-L. Kniep, M. Tovar, R. W. [74] U. Simon, S. Arndt, T. Otremba, T. Schlingmann, O. Goerke,
Fischer, J. K. Nørskov, R. Schlçgl, Science 2012, 336, 893 – 897. K. P. Dinse, R. Schomaecker, H. Schubert, Catal. Commun.
[57] a) C. L. Zhao, I. E. Wachs, J. Phys. Chem. C 2008, 112, 11363 – 2012, 18, 132 – 136.
11372; b) A. Gurlo, R. Riedel, Angew. Chem. Int. Ed. 2007, 46, [75] U. Zavyalova, M. Geske, R. Horn, G. Weinberg, W. Frandsen,
3826 – 3848; Angew. Chem. 2007, 119, 3900 – 3923. M. Schuster, R. Schlçgl, ChemCatChem 2011, 3, 949 – 959.
[58] a) R. L pez-Medina, H. Golinska, M. Ziolek, M. O. Guerrero- [76] L. Andrews, J. Chem. Phys. 1969, 50, 4288.
Perez, M. A. Banares, Catal. Today 2010, 158, 139 – 145; b) I. E. [77] H. Kudo, C. H. Wu, H. R. Ihle, J. Nucl. Mater. 1978, 78, 389.
Wachs, J. M. Jehng, W. Ueda, J. Phys. Chem. B 1994, 109, 2275 – [78] N. A. Richter, S. Sicolo, S. V. Levchenko, J. Sauer, M. Scheffler,
2284; c) I. E. Wachs, Catal. Today 2005, 100, 79 – 94; d) C. Phys. Rev. Lett. 2013, 111, 045502.
Sievers, S. R. Bare, E. Stavitski, Catal. Today 2013, 205, 1 – 2;

3510 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

[79] K. Kwapien, J. Paier, J. Sauer, M. Geske, U. Zavyalova, R. d) H. U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner,
Horn, P. Schwach, A. Trunschke, R. Schlçgl, Angew. Chem. Int. M. Studer, Adv. Synth. Catal. 2003, 345, 103 – 151.
Ed. 2014, 53, 8774 – 8778; Angew. Chem. 2014, 126, 8919 – 8923. [100] I. N. Chorkendorff, J. W. Niemantsverdriet, Concepts of
[80] a) M. Sterrer, E. Fischbach, T. Risse, H. J. Freund, Phys. Rev. Modern Catalysis and Kinetics, Wiley-VCH, Weinheim, 2007.
Lett. 2005, 94, 186101; b) M. Anpo, M. Che, B. Fubini, E. [101] a) P. Stoltze, J. K. Nørskov, H. Topsøe, M. Boudart, The Surface
Garrone, E. Giamello, M. C. Paganini, Top. Catal. 1999, 8, 189 – Science Based Ammonia Kinetics Revisited, J. C. Baltzer AG,
198; c) P. Schwach, M. G. Willinger, A. Trunschke, R. Schlçgl, 1994; b) B. Hammer, J. K. Nørskov in Advances in Catalysis,
Angew. Chem. Int. Ed. 2013, 52, 11381 – 11384; Angew. Chem. Vol. 45 (Eds.: B. C. Gates, H. Knçzinger), 2000, pp. 71 – 129;
2013, 125, 11591 – 11594. c) G. Djega-Mariadassou, M. Boudart, J. Catal. 2003, 216, 89 –
[81] P. T. Schwach, W. Frandsen, M. Willinger, R. Schlçgl, A. 97; d) M. Boudart, Chem. Rev. 1995, 95, 661 – 666.
Trunschke, J. Catal. 2015, submitted. [102] H. S. Taylor, Proc. R. Soc. London Ser. A 1925, 108, 105 – 111.
[82] J. K. Nørskov, T. Bligaard, A. Logadottir, S. Bahn, L. B. [103] R. Schlçgl, Adv. Catal. 2008, 52, 273 – 338.
Hansen, M. Bollinger, H. Bengaard, B. Hammer, Z. Sljivanca- [104] H. S. Taylor, J. Phys. Chem. 1926, 30, 145 – 171.
nin, M. Mavrikakis, Y. Xu, S. Dahl, C. J. H. Jacobsen, J. Catal. [105] a) M. Behrens, A. Furche, I. Kasatkin, A. Trunschke, W. Busser,
2002, 209, 275 – 278. M. Muhler, B. Kniep, R. Fischer, R. Schlçgl, ChemCatChem
[83] a) D. W. Goodman, Surf. Sci. 1994, 299, 837 – 848; b) J. A. 2010, 2, 816 – 818; b) I. Kasatkin, P. Kurr, B. Kniep, A.
Rodriguez, D. W. Goodman, Surf. Sci. Rep. 1991, 14, 1 – 107; Trunschke, R. Schlçgl, Angew. Chem. Int. Ed. 2007, 46, 7324 –
c) I. B. Parker, K. C. Waugh, M. Bowker, J. Catal. 1988, 114, 7327; Angew. Chem. 2007, 119, 7465 – 7468.
457 – 459; d) J. Dumesic, H. Topsøe, S. Khammouma, M. [106] T. Kandemir, F. Girgsdies, T. C. Hansen, K.-D. Liss, I. Kasatkin,
Boudart, J. Catal. 1975, 37, 503 – 512. E. L. Kunkes, G. Wowsnick, N. Jacobsen, R. Schlçgl, M.
[84] G. Spoto, E. N. Gribov, G. Ricchiardi, A. Damin, D. Scarano, S. Behrens, Angew. Chem. Int. Ed. 2013, 52, 5166 – 5170; Angew.
Bordiga, C. Lamberti, A. Zecchina, Prog. Surf. Sci. 2004, 76, Chem. 2013, 125, 5271 – 5276.
71 – 146. [107] B. E. Warren, Prog. Met. Phys. 1959, 8, 147 – 202.
[85] F. Finocchi, P. Geysermans, A. Bourgeois, Phys. Chem. Chem. [108] a) T. Kandemir, I. Kasatkin, F. Girgsdies, S. Zander, S. Khl, M.
Phys. 2012, 14, 13692 – 13701. Tovar, R. Schlçgl, M. Behrens, Top. Catal. 2014, 57, 188 – 206;
[86] U. Zavyalova, M. Geske, R. Horn, G. Weinberg, W. Frandsen, b) T. Kandemir, D. Wallacher, T. Hansen, K. D. Liss, R.
M. Schuster, R. Schlçgl, ChemCatChem 2011, 3, 949 – 959. Naumann dAlnoncourt, R. Schlçgl, M. Behrens, Nucl. Instrum.
[87] P. Geysermans, F. Finocchi, J. Goniakowski, R. Hacquart, J. Methods Phys. Res. Sect. A 2012, 673, 51 – 55.
Jupille, Phys. Chem. Chem. Phys. 2009, 11, 2228 – 2233. [109] a) M. Behrens, R. Schlçgl, Z. Anorg. Allg. Chem. 2013, 639,
[88] J. S. J. Hargreaves, G. J. Hutchings, R. W. Joyner, C. J. Kiely, J. 2683 – 2695; b) M. Behrens, S. Zander, P. Kurr, N. Jacobsen, J.
Catal. 1992, 135, 576 – 595. Senker, G. Koch, T. Ressler, R. W. Fischer, R. Schlçgl, J. Am.
[89] H. P. Boehm, Angew. Chem. Int. Ed. Engl. 1966, 5, 533 – 544; Chem. Soc. 2013, 135, 6061 – 6068.
Angew. Chem. 1966, 78, 617 – 628. [110] a) J. W. Evans, M. S. Wainwright, A. J. Bridgewater, D. J.
[90] J. T. Newberg, D. E. Starr, S. Yamamoto, S. Kaya, T. Kende- Young, Appl. Catal. 1983, 7, 75 – 83; b) M. Muhler, L. P.
lewicz, E. R. Mysak, S. Porsgaard, M. B. Salmeron, G. E. Nielsen, E. Tçrnqvist, B. S. Clausen, H. Topsøe, Catal. Lett.
Brown, A. Nisson, H. Bluhm, J. Phys. Chem. C 2011, 115, 1992, 14, 241 – 249; c) J. J. Scholten, Ja. Konvalin, Trans. Fara-
12864 – 12872. day Soc. 1969, 65, 2465.
[91] R. Włodarczyk, M. Sierka, K. Kwapień, J. Sauer, E. Carrasco, [111] a) W. D. Michalak, G. A. Somorjai, Top. Catal. 2013, 56, 1611 –
A. Aumer, J. F. Gomes, M. Sterrer, H. J. Freund, J. Phys. Chem. 1622; b) G. Ertl, Surf. Sci. 1994, 299, 742 – 754.
C 2011, 115, 6764 – 6774. [112] a) M. Boudart, J. Mol. Catal. 1985, 30, 27 – 38; b) M. Boudart,
[92] a) G. A. Somorjai, Surf. Sci. 1994, 299, 849 – 866; b) D. R. H. Topsøe, M. Boudart, J. K. Nørskov, Ammonia Synthesis: The
Strongin, J. Carrazza, S. R. Bare, G. A. Somorjai, J. Catal. 1987, Bellwether Reaction in Heterogeneous Catalysis, J. C. Baltzer
103, 213 – 215; c) L. A. West, G. A. Somorjai, J. Chem. Phys. AG, 1994.
1971, 54, 2864; d) S. Dahl, P. A. Taylor, E. Tçrnqvist, I. [113] a) G. C. Bond, S. N. Namijo, J. Catal. 1989, 118, 507 – 510;
Chorkendorff, J. Catal. 1998, 178, 679 – 686. b) Ra. Dallabet, J. Catal. 1974, 34, 57 – 60; c) J. J. F. Scholten, A.
[93] Y. Cui, X. Shao, M. Baldofski, J. Sauer, N. Nilius, H. J. Freund, Vanmontfoort, J. Catal. 1962, 1, 85 – 92; d) A. Gervasini, S.
Angew. Chem. Int. Ed. 2013, 52, 11385 – 11387; Angew. Chem. Bennici, Appl. Catal. A: Gen. 2005, 281, 199 – 205.
2013, 125, 11595 – 11598. [114] A. Pozio, M. De Francesco, A. Cemmi, F. Cardellini, L. Giorgi,
[94] H. Metiu, S. Chretien, Z. P. Hu, B. Li, X. Y. Sun, J. Phys. Chem. J. Power Sources 2002, 105, 13 – 19.
C 2012, 116, 10439 – 10450. [115] A. A. Lizzio, L. R. Radovic, Ind. Eng. Chem. Res. 1991, 30,
[95] a) Y. Martynova, S. Shaikhutdinov, H. J. Freund, Chem- 1735 – 1744.
CatChem 2013, 5, 2162 – 2166; b) A. R. Derk, B. Li, S. [116] K. Amakawa, S. Wrabetz, J. Krçhnert, G. Tzolova-Mller, R.
Sharma, G. M. Moore, E. W. McFarland, H. Metiu, Catal. Schlçgl, A. Trunschke, J. Am. Chem. Soc. 2012, 134, 11462 –
Lett. 2013, 143, 406 – 410; c) V. Shapovalov, H. Metiu, J. Catal. 11473.
2007, 245, 205 – 214. [117] K. Amakawa, L. Sun, C. Guo, M. Hvecker, P. Kube, I. E.
[96] S. Dahl, A. Logadottir, R. C. Egeberg, J. H. Larsen, I. Wachs, S. Lwin, A. I. Frenkel, A. Patlolla, K. Hermann, R.
Chorkendorff, E. Tornqvist, J. K. Norskov, Phys. Rev. Lett. Schlçgl, A. Trunschke, Angew. Chem. Int. Ed. 2013, 52, 13553 –
1999, 83, 1814 – 1817. 13557; Angew. Chem. 2013, 125, 13796 – 13800.
[97] a) G. Ertl, Angew. Chem. Int. Ed. Engl. 1990, 29, 1219 – 1227; [118] a) G. Ertl, Catal. Rev. Sci. Eng. 1980, 21, 201 – 223; b) G. Ertl,
Angew. Chem. 1990, 102, 1258 – 1266; b) G. A. Somorjai, H. J. R. Jennings, Elementary Steps in Ammonia Synthesis: The
Frei, J. Y. Park, J. Am. Chem. Soc. 2009, 131, 16589 – 16605. Surf. Sci. Approach, Plenum, New York, 1991; c) G. A.
[98] R. A. Sheldon, Green Chem. 2007, 9, 1273 – 1283. Somorjai, R. M. Rioux, Catal. Today 2005, 100, 201 – 215.
[99] a) M. Beller, B. Cornils, C. D. Frohning, C. W. Kohlpaintner, J. [119] I. Langmuir, Trans. Faraday Soc. 1922, 17, 0621 – 0654.
Mol. Catal. A 1995, 104, 17 – 85; b) M. Peters, B. Koehler, W. [120] R. P. Donnelly, C. N. Hinshelwood, J. Chem. Soc. 1929, 1727 –
Kuckshinrichs, W. Leitner, P. Markewitz, T. E. Mueller, Chem- 1733.
SusChem 2011, 4, 1216 – 1240; c) M. S. Sanford, J. A. Love, [121] A. K. Engstfeld, S. Brimaud, R. J. Behm, Angew. Chem. Int. Ed.
R. H. Grubbs, J. Am. Chem. Soc. 2001, 123, 6543 – 6554; 2014, 53, 12936 – 12940; Angew. Chem. 2014, 126, 13150 – 13154.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3511
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

[122] X. Bao, G. Lehmpfuhl, G. Weinberg, R. Schlçgl, G. Ertl, J. pp. 108 – 120; c) R. Schlçgl, C. Hess in Nanostructured Cata-
Chem. Soc. Faraday Trans. 1992, 88, 865 – 872. lysts: Selective Oxidations (Eds.: C. Hess, R. Schlçgl), The
[123] M. Baerns, R. Imbihl, V. A. Kondratenko, R. Kraehnert, W. K. Royal Society of Chemistry, 2011, pp. 355 – 397.
Offermans, R. A. van Santen, A. Scheibe, J. Catal. 2005, 232, [141] P. Mars, D. W. Van Krevelen, Chem. Eng. Sci. 1954, 3, 41.
226 – 238. [142] G. Ertl, M. Huber, N. Thiele, Z. Naturforsch. A 1978, 34, 30 – 39.
[124] a) H. J. Grabke, Z. Phys. Chem. Neue Folge 1976, 100, 185 – [143] a) G. Ertl, Angew. Chem. Int. Ed. 2013, 52, 52 – 60; Angew.
200; b) H. J. Grabke, Ber. Bunsen-Ges. 1968, 72, 533 – 541; c) A. Chem. 2013, 125, 52 – 61; b) I. E. Wachs, R. J. Madix, J. Catal.
Colombo, Anal. Chim. Acta 1976, 81, 397 – 407; d) W. G. 1978, 53, 208 – 227; c) M. Bowker, M. A. Barteau, R. J. Madix,
Guldner, Talanta 1961, 8, 191 – 202. Surf. Sci. 1980, 92, 528 – 548; d) M. A. Barteau, E. I. Ko, R. J.
[125] D. Vogel, Z. Budinska, C. Spiel, R. Schloegl, Y. Suchorski, G. Madix, Surf. Sci. 1981, 102, 99 – 117.
Rupprechter, Catal. Lett. 2013, 143, 235 – 240. [144] J. Libuda, I. Meusel, J. Hoffmann, J. Hartmann, L. Piccolo,
[126] a) X. Bao, G. Lehmpfuhl, G. Weinberg, R. Schlçgl, G. Ertl, J. C. R. Henry, H. J. Freund, J. Chem. Phys. 2001, 114, 4669 – 4684.
Chem. Soc. Faraday Trans. 1992, 88, 865 – 872; b) A. J. Nagy, G. [145] J. H. Larsen, I. Chorkendorff, Surf. Sci. Rep. 1999, 35, 165 – 222.
Mestl, D. Herein, G. Weinberg, E. Kitzelmann, R. Schlçgl, J. [146] Y. Yang, C. A. Mims, D. H. Mei, C. H. F. Peden, C. T. Camp-
Catal. 1999, 182, 417 – 429. bell, J. Catal. 2013, 298, 10 – 17.
[127] G. Ertl, J. Vac. Sci. Technol. A 1983, 1, 1247 – 1253. [147] L. Vattuone, Y. Y. Yeo, R. Kose, D. A. King, Surf. Sci. 2000, 447,
[128] a) P. Stoltze, J. K. Norskov, J. Catal. 1988, 110, 1 – 10; b) H. D. 1 – 14.
Vandervell, K. C. Waugh, Chem. Phys. Lett. 1990, 171, 462 – [148] a) J. H. Fischer-Wolfarth, J. A. Farmer, J. M. Flores-Camacho,
468; c) M. Bowker, I. Parker, K. C. Waugh, Surf. Sci. 1988, 197, A. Genest, I. V. Yudanov, N. Rosch, C. T. Campbell, S.
L223 – L229. Schauermann, H. J. Freund, Phys. Rev. B 2010, 81, 241416;
[129] P. Stoltze, J. K. Norskov, Top. Catal. 1994, 1, 253 – 263. b) Y. N. Sun, Z. H. Qin, M. Lewandowski, S. Shaikhutdinov,
[130] a) A. Hellman, E. J. Baerends, M. Biczysko, T. Bligaard, C. H. H. J. Freund, Surf. Sci. 2009, 603, 3099 – 3103.
Christensen, D. C. Clary, S. Dahl, R. van Harrevelt, K. Hon- [149] S. Wrabetz, X. B. Yang, G. Tzolova-Mller, R. Schlçgl, F. C.
kala, H. Jonsson, G. J. Kroes, M. Luppi, U. Manthe, J. K. Jentoft, J. Catal. 2010, 269, 351 – 358.
Norskov, R. A. Olsen, J. Rossmeisl, E. Skulason, C. S. Tauter- [150] a) R. Arrigo, S. Wrabetz, M. E. Schuster, D. Wang, A. Villa, D.
mann, A. J. C. Varandas, J. K. Vincent, J. Phys. Chem. B 2006, Rosenthal, F. Girsgdies, G. Weinberg, L. Prati, R. Schlçgl, D. S.
110, 17719 – 17735; b) G. Ertl, I. T. Horvçth, E. Iglesia, M. T. Su, Phys. Chem. Chem. Phys. 2012, 14, 10523 – 10532; b) K.
Klein, J. A. Lercher, A. J. Russel, E. I. Stiefel in Encyclopedia
Amakawa, S. Wrabetz, J. Krçhnert, G. Tzolova-Mller, R.
of Catalysis, Wiley, New York, 2002.
Schlçgl, A. Trunschke, J. Am. Chem. Soc. 2012, 134, 11462 –
[131] G. Ertl, Z. Anorg. Allg. Chem. 2012, 638, 487 – 489.
11473; c) H. A. Prescott, Z.-J. Li, E. Kemnitz, A. Trunschke, J.
[132] a) S. Vçlkening, K. Bedurftig, K. Jacobi, J. Wintterlin, G. Ertl,
Deutsch, H. Lieske, A. Auroux, J. Catal. 2005, 234, 119 – 130;
Phys. Rev. Lett. 1999, 83, 2672 – 2675; b) J. Wintterlin, S.
d) M. V. Lopez-Ramon, F. Stoeckli, C. Moreno-Castilla, F.
Vçlkening, T. V. W. Janssens, T. Zambelli, G. Ertl, Science
Carrasco-Marin, Carbon 1999, 37, 1215 – 1221; e) J. Le Bars,
1997, 278, 1931 – 1934; c) T. Zambelli, J. Wintterlin, J. Trost, G.
J. C. Vedrine, A. Auroux, S. Trautmann, M. Baerns, Appl. Catal.
Ertl, Science 1996, 273, 1688 – 1690; d) D. Vogel, C. Spiel, Y.
A 1994, 119, 341 – 354.
Suchorski, A. Urich, R. Schlçgl, G. Rupprechter, Surf. Sci. 2011,
[151] H. Petitjean, K. Tarasov, F. Delbecq, P. Sautet, J. M. Krafftt, P.
605, 1999 – 2005; e) Y. Suchorski, C. Spiel, D. Vogel, W.
Bazin, M. C. Paganini, E. Giamello, M. Che, H. Lauron-Pernot,
Drachsel, R. Schlçgl, G. Rupprechtert, ChemPhysChem 2010,
G. Costentin, J. Phys. Chem. C 2010, 114, 3008 – 3016.
11, 3231 – 3235.
[152] a) J. Liu, F. Mohamed, J. Sauer, J. Catal. 2014, 317, 75 – 82;
[133] a) G. A. Somorjai, Y. Li, Introduction to Surface Chemistry and
b) H. J. Freund, G. Meijer, M. Scheffler, R. Schlçgl, M. Wolf,
Catalysis, Wiley, Hoboken, 2010; b) G. A. Somorjai, Langmuir
1991, 7, 3176 – 3182; c) D. R. Strongin, G. A. Somorjai, J. Catal. Angew. Chem. Int. Ed. 2011, 50, 10064 – 10094; Angew. Chem.
1989, 118, 99 – 110; d) D. R. Strongin, S. R. Bare, G. A. 2011, 123, 10242 – 10275.
Somorjai, J. Catal. 1987, 103, 289 – 301. [153] a) B. L. M. Hendriksen, J. W. M. Frenken, Phys. Rev. Lett. 2002,
[134] H. Lynggaard, A. Andreasen, C. Stegelmann, P. Stoltze, Prog. 89, 046101; b) P. B. Rasmussen, B. L. M. Hendriksen, H. Zeijle-
Surf. Sci. 2004, 77, 71 – 137. maker, H. G. Ficke, J. W. M. Frenken, Rev. Sci. Instrum. 1998,
[135] C. Spiel, D. Vogel, Y. Suchorski, W. Drachsel, R. Schlçgl, G. 69, 3879 – 3884.
Rupprechter, Catal. Lett. 2011, 141, 625 – 632. [154] M. Geske, O. Korup, R. Horn, Catal. Sci. Technol. 2013, 3, 169 –
[136] a) R. Lloyd, T. W. Hansen, W. Ranke, F. C. Jentoft, R. Schlçgl, 175.
Appl. Catal. A 2011, 391, 215 – 224; b) S. van Donk, J. H. Bitter, [155] a) E. V. Kondratenko, N. Steinfeldt, M. Baerns, Phys. Chem.
K. P. de Jong, Appl. Catal. A 2001, 212, 97 – 116; c) B. S. Klose, Chem. Phys. 2006, 8, 1624 – 1633; b) H. Zanthoff, M. Baerns,
F. C. Jentoft, R. Schlçgl, J. Catal. 2005, 233, 68 – 80; d) A. H. P. Ind. Eng. Chem. Res. 1990, 29, 2 – 10.
Hahn, R. E. Jentoft, T. Ressler, G. Weinberg, R. Schlçgl, F. C. [156] a) S. Polarz, J. Strunk, V. Ischenko, M. W. E. van den Berg, O.
Jentoft, J. Catal. 2005, 236, 324 – 334; e) F. C. Jentoft, A. Hahn, Hinrichsen, M. Muhler, M. Driess, Angew. Chem. Int. Ed. 2006,
J. Krçhnert, G. Lorenz, R. E. Jentoft, T. Ressler, U. Wild, R. 45, 2965 – 2969; Angew. Chem. 2006, 118, 3031 – 3035; b) H.
Schlçgl, C. Hßner, K. Kçhler, J. Catal. 2004, 224, 124 – 137; Madhavaram, H. Idriss, S. Wendt, Y. D. Kim, M. Knapp, H.
f) R. Ahmad, J. Melsheimer, F. C. Jentoft, R. Schlçgl, J. Catal. Over, J. Assmann, E. Loffler, M. Muhler, J. Catal. 2001, 202,
2003, 218, 365 – 374. 296 – 307; c) H. Bielawa, O. Hinrichsen, A. Birkner, M. Muhler,
[137] J. M. Basset, A. Choplin, J. Mol. Catal. 1983, 21, 95 – 108. Angew. Chem. Int. Ed. 2001, 40, 1061 – 1063; Angew. Chem.
[138] R. Schlçgl, A. Knop-Gericke, M. Hvecker, U. Wild, D. Frickel, 2001, 113, 1093 – 1096; d) X. Bao, M. Muhler, B. Pettinger, R.
T. Ressler, R. E. Jentoft, J. Wienold, A. Blume, G. Mestl, O. Schlçgl, G. Ertl, Catal. Lett. 1993, 22, 215 – 225; e) J. P.
Timpe, Y. Uchida, Top. Catal. 2001, 15, 219 – 228. Tessonnier, M. Becker, W. Xia, F. Girgsdies, R. Blume, L. D.
[139] B. Frank, R. Fortrie, C. Hess, R. Schlçgl, R. Schomcker, Appl. Yao, D. S. Su, M. Muhler, R. Schlçgl, ChemCatChem 2010, 2,
Catal. A 2009, 353, 288 – 295. 1559 – 1561; f) S. Kundu, Y. M. Wang, W. Xia, M. Muhler, J.
[140] a) R. Schlçgl, Top. Catal. 2011, 54, 627 – 638; b) R. Schlçgl, Phys. Chem. C 2008, 112, 16869 – 16878; g) M. Guraya, S.
Catalysis in Application (Eds.: S. D. Jackson, J. S. J. Hargreaves, Sprenger, W. Rarog-Pilecka, D. Szmigiel, Z. Kowalczyk, M.
D. Lennon), The Royal Society of Chemistry, Glasgow, 2003, Muhler, Appl. Surf. Sci. 2004, 238, 77 – 81; h) J. Assmann, V.

3512 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

Narkhede, L. Khodeir, E. Loffler, O. Hinrichsen, A. Birkner, H. [171] a) M. F. R. Pereira, J. L. Figueiredo, J. J. M. Orfao, P. Serp, P.
Over, M. Muhler, J. Phys. Chem. B 2004, 108, 14634 – 14642. Kalck, Y. Kihn, Carbon 2004, 42, 2807 – 2813; b) M. F. R.
[157] G. Emig, M. A. Liauw, Top. Catal. 2002, 21, 11 – 24. Pereira, J. J. M. Orfao, J. L. Figueiredo, Appl. Catal. A 2001,
[158] a) H. H. Rotermund, G. Has, R. U. Franz, R. Tromp, G. Ertl, 218, 307 – 318; c) G. E. Vrieland, P. G. Menon, Appl. Catal.
Science 1995, 270, 608 – 610; b) R. Imbihl, G. Ertl, Chem. Rev. 1991, 77, 1 – 8; d) R. Fiedorow, W. Przystajko, M. Sopa, J. Catal.
1995, 95, 697 – 733. 1981, 68, 33 – 41.
[159] a) A. Boubnov, A. Ganzler, S. Conrad, M. Casapu, J. D. [172] a) J. J. Delgado, R. Vieira, G. Rebmann, D. S. Su, N. Keller,
Grunwaldt, Top. Catal. 2013, 56, 333 – 338; b) D. Rosenthal, M. J. Ledoux, R. Schlçgl, Carbon 2006, 44, 809 – 812; b) J. J.
Phys. Status Solidi A 2011, 208, 1217 – 1222. Delgado, D. S. Su, G. Rebmann, N. Keller, A. Gajovic, R.
[160] a) I. Bçttger, T. Schedel-Niedrig, O. Timpe, R. Gottschall, M. Schlçgl, J. Catal. 2006, 244, 126 – 129; c) N. Keller, N. I.
Hvecker, T. Ressler, R. Schlçgl, Chem. Eur. J. 2000, 6, 1870 – Maksimova, V. V. Roddatis, M. Schur, G. Mestl, Y. V. Butenko,
1876; b) V. V. Kaichev, A. Y. Gladky, I. P. Prosvirin, A. A. V. L. Kuznetsov, R. Schlçgl, Angew. Chem. Int. Ed. 2002, 41,
Saraev, M. Hvecker, A. Knop-Gericke, R. Schlçgl, V. I. 1885 – 1888; Angew. Chem. 2002, 114, 1962 – 1966; d) G. Mestl,
Bukhtiyarov, Surf. Sci. 2013, 609, 113 – 118. N. I. Maksimova, N. Keller, V. V. Roddatis, R. Schlçgl, Angew.
[161] a) O. Hinrichsen, F. Rosowski, A. Hornung, M. Muhler, G. Ertl, Chem. Int. Ed. 2001, 40, 2066 – 2068; Angew. Chem. 2001, 113,
J. Catal. 1997, 165, 33 – 44; b) F. Rosowski, O. Hinrichsen, M. 2122 – 2125.
Muhler, G. Ertl, Catal. Lett. 1996, 36, 229 – 235; c) P. Stoltze, H. [173] R. Schlçgl in Advances in Catalysis, Vol. 56, (Eds.: B. C. Gates,
Topsoe, A. Nielsen, Structure an Surface Chemistry of Industrial H. Knçzinger), Elsevier, London, 2013, pp. 103 – 185.
Ammonia Synthesis Catalysts, Springer, Berlin, 1995; d) M. [174] O. Shekhah, W. Ranke, A. Schle, G. Kolios, R. Schlçgl, Angew.
Muhler, F. Rosowski, G. Ertl, Catal. Lett. 1994, 24, 317 – 331. Chem. Int. Ed. 2003, 42, 5760 – 5763; Angew. Chem. 2003, 115,
[162] a) Y. M. Jin, A. K. Datye, E. Rightor, R. Gulotty, W. Water- 5938 – 5941.
man, M. Smith, M. Holbrook, J. Maj, J. Blackson, J. Catal. 2001, [175] Y. Joseph, M. Whn, A. Niklewski, W. Ranke, W. Weiss, C.
203, 292 – 306; b) D. Gçbke, Y. Romanyshyn, S. Guimond, J.-M. Wçll, R. Schlçgl, Phys. Chem. Chem. Phys. 2000, 2, 5314 – 5319.
Sturm, H. Kuhlenbeck, J. Doebler, U. Reinhardt, M.-V. [176] a) J. Zhang, D. S. Su, R. Blume, R. Schlçgl, R. Wang, X. Yang,
Ganduglia-Pirovano, J. Sauer, H.-J. Freund, Angew. Chem. A. Gajović, Angew. Chem. Int. Ed. 2010, 49, 8640 – 8644;
Int. Ed. 2009, 48, 3695 – 3698; Angew. Chem. 2009, 121, 3750 – Angew. Chem. 2010, 122, 8822 – 8826; b) E. P. J. Parrott, J. A.
3753; c) N. Magg, B. Immaraporn, J. B. Giorgi, T. Schroeder, M. Zeitler, J. McGregor, S.-P. Oei, S.-C. Tan, W. I. Milne, J.-P.
Bumer, J. Dçbler, Z. Wu, E. Kondratenko, M. Cherian, M. Tessonnier, D. S. Su, R. Schlçgl, L. F. Gladden, J. Phys. Chem. C
2009, 113, 10554 – 10559; c) J. A. Maci-Agull , D. Cazorla-
Baerns, P. C. Stair, J. Sauer, H.-J. Freund, J. Catal. 2004, 226,
Amor s, A. Lineares-Solano, U. Wild, D. S. Su, R. Schlçgl,
88 – 100; d) H. J. Freund, Angew. Chem. Int. Ed. Engl. 1997, 36,
Catal. Today 2005, 102 – 103, 248 – 253.
452 – 475; Angew. Chem. 1997, 109, 444 – 468.
[177] A. Schle, O. Shekhah, W. Ranke, R. Schlçgl, G. Kolios, J.
[163] a) F. Tao, S. Dag, L. W. Wang, Z. Liu, D. R. Butcher, H. Bluhm,
Catal. 2005, 231, 173 – 180.
M. Salmeron, G. A. Somorjai, Science 2010, 327, 850 – 853;
[178] a) P. Sautet, Actual. Chim. 2002, 73 – 75; b) M. Neurock, D. H.
b) R. Hggblad, J. B. Wagner, S. Hansen, A. Andersson, J.
Mei, Top. Catal. 2002, 20, 5 – 23.
Catal. 2008, 258, 345 – 355.
[179] a) E. M. Vass, M. Hvecker, S. Zafeiratos, D. Teschner, A.
[164] a) M. Muhler, R. Schlçgl, G. Ertl, J. Catal. 1992, 138, 413 – 444;
Knop-Gericke, R. Schlçgl, J. Phys. Condens. Matter 2008, 20,
b) M. Muhler, J. Schtze, M. Wesemann, T. Rayment, A. Dent,
184016; b) P. Kampe, L. Giebeler, D. Samuelis, J. Kunert, A.
R. Schlçgl, G. Ertl, J. Catal. 1990, 126, 339 – 360; c) M. Muhler,
Drochner, F. Haass, A. H. Adams, J. Ott, S. Endres, G.
R. Schlçgl, A. Reller, G. Ertl, Catal. Lett. 1989, 2, 201 – 210;
Schimanke, T. Buhrmester, M. Martin, H. Fuess, H. Vogel,
d) M. Muhler, R. Schlçgl, G. Ertl, Surf. Interface Anal. 1988, 12,
Phys. Chem. Chem. Phys. 2007, 9, 3577 – 3589; c) R. Blume, M.
233 – 238.
Hvecker, S. Zafeiratos, D. Teschner, E. Vass, P. Schnoerch, A.
[165] a) A. Schraut, G. Emig, H. G. Sockel, Appl. Catal. 1987, 29,
Knop-Gericke, R. Schloegl, S. Lizzit, P. Dudin, A. Barinov, M.
311 – 326; b) G. Emig, H. Hofmann, J. Catal. 1983, 84, 15 – 26.
Kiskinova, Phys. Chem. Chem. Phys. 2007, 9, 3648 – 3657; d) R.
[166] a) C. Kuhrs, Y. Arita, W. Weiss, W. Ranke, R. Schlçgl, Top. Blume, M. Hvecker, S. Zafeiratos, D. Teschner, A. Knop-
Catal. 2001, 14, 111 – 123; b) G. Ketteler, W. Weiss, W. Ranke, Gericke, R. Schlçgl, P. Dudin, A. Barinov, M. Kiskinova, Catal.
R. Schlçgl, Phys. Chem. Chem. Phys. 2001, 3, 1114 – 1122; c) Y. Today 2007, 124, 71 – 79; e) M. Hvecker, A. Knop-Gericke, H.
Joseph, G. Ketteler, C. Kuhrs, W. Ranke, W. Weiss, R. Schlçgl, Bluhm, E. Kleimenov, W. Mayer, M. Fait, R. Schlçgl, Appl.
Phys. Chem. Chem. Phys. 2001, 3, 4141 – 4153; d) W. Weiss, R. Surf. Sci. 2004, 230, 272 – 282; f) E. Savinova, A. Scheybal, M.
Schlçgl, Top. Catal. 2000, 13, 75 – 90; e) S. K. Shaikhutdinov, W. Danckwerts, U. Wild, B. Pettinger, K. Doblhofer, R. Schl gl, G.
Weiss, R. Schlçgl, Appl. Surf. Sci. 2000, 161, 497 – 507; f) D. Ertl, Faraday Discuss. 2002, 121, 181 – 198; g) T. Ressler, O.
Zscherpel, W. Ranke, W. Weiss, R. Schlçgl, J. Chem. Phys. 1998, Timpe, T. Neisius, J. Find, G. Mestl, M. Dieterle, R. Schlçgl, J.
108, 9506 – 9515; g) W. Weiss, D. Zscherpel, R. Schlçgl, Catal. Catal. 2000, 191, 75 – 85; h) B. Grzybowska-Świerkosz, Top.
Lett. 1998, 52, 215 – 220; h) W. Weiss, M. Ritter, D. Zscherpel, Catal. 2000, 11 – 12, 23 – 42; i) M. A. BaÇares, J. H. Cardoso, F.
M. Swoboda, R. Schlçgl, J. Vac. Sci. Technol. A 1998, 16, 21 – 29; Agullo-Rueda, J. M. Correa-Bueno, J. L. G. Fierro, Catal. Lett.
i) X. G. Wang, W. Weiss, S. K. Shaikhutdinov, M. Ritter, M. 2000, 64, 191 – 196; j) I. Wachs, Top. Catal. 1999, 8, 57 – 63; k) A.
Petersen, F. Wagner, R. Schlçgl, M. Scheffler, Phys. Rev. Lett. Nagy, G. Mestl, T. Rhle, G. Weinberg, R. Schlçgl, J. Catal.
1998, 81, 1038 – 1041. 1998, 179, 548 – 559; l) J. M. Jehng, J. Phys. Chem. B 1998, 102,
[167] W. Weiss, W. Ranke, Prog. Surf. Sci. 2002, 70, 1 – 151. 5816 – 5822; m) J.-C. Volta, Catal. Today 1996, 32, 29 – 36; n) J.-
[168] D. Zscherpel, W. Weiss, R. Schlçgl, Surf. Sci. 1997, 382, 326 – M. Jehng, H. Hu, X. Gao, I. E. Wachs, Catal. Today 1996, 28,
335. 335 – 350.
[169] a) W. Weiss, R. Schlçgl, Top. Catal. 2000, 13, 75 – 90; b) W. [180] a) E. V. Kondratenko, M. Cherian, M. Baerns, D. S. Su, R.
Weiss, M. Ritter, D. Zscherpel, M. Swoboda, R. Schlçgl, J. Vac. Schlçgl, X. Wang, I. E. Wachs, J. Catal. 2005, 234, 131 – 142; b) J.
Sci. Technol. A 1998, 16, 21 – 29. Haber, W. Turek, J. Catal. 2000, 190, 320 – 326; c) V. V. Guliants,
[170] a) A. Schle, U. Nieken, O. Shekhah, W. Ranke, R. Schlçgl, G. H. H. Brongersma, A. Knoester, A. M. Gaffney, S. Han, Top.
Kolios, Phys. Chem. Chem. Phys. 2007, 9, 3619 – 3634; b) O. Catal. 2006, 38, 41 – 50; d) V. Guliants Vadim, R. Bhandari, H.
Shekhah, W. Ranke, R. Schlçgl, J. Catal. 2004, 225, 56 – 68. Brongersma Hidde, A. Knoester, M. Gaffney Anne, S. Han, J.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3513
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Phys. Chem. B 2005, 109, 10234 – 10242; e) M. A. Carreon, V. V. Angew. Chem. 2009, 121, 8150 – 8153; c) Q. Guo, S. Lee, D. W.
Guliants, Catal. Today 2003, 78, 303 – 310. Goodman, Surf. Sci. 1999, 437, 38 – 48.
[181] a) G. Rupprechter, M. Morkel, H. J. Freund, R. Hirschl, Surf. [192] G. Ertl, H. J. Freund, Phys. Today 1999, 52, 32 – 38.
Sci. 2004, 554, 43 – 59; b) H. J. Freund, M. Baumer, J. Libuda, T. [193] Q. L. Tang, Q. J. Hong, Z. P. Liu, J. Catal. 2009, 263, 114 – 122.
Risse, G. Rupprechter, S. Shaikhutdinov, J. Catal. 2003, 216, [194] a) C. Heine, M. Havecker, E. Stotz, F. Rosowski, A. Knop-
223 – 235; c) A. M. Doyle, S. K. Shaikhutdinov, S. D. Jackson, Gericke, A. Trunschke, M. Eichelbaum, R. Schlçgl, J. Phys.
H. J. Freund, Angew. Chem. Int. Ed. 2003, 42, 5240 – 5243; Chem. C 2014, 118, 20405 – 20412; b) K. R. Asmis, T. Wende,
Angew. Chem. 2003, 115, 5398 – 5401; d) H. J. Freund, N. Ernst, M. Brummer, O. Gause, G. Santambrogio, E. C. Stanca-
T. Risse, H. Hamann, G. Rupprechter, Phys. Status Solidi A Kaposta, J. Dobler, A. Niedziela, J. Sauer, Phys. Chem.
2001, 187, 257 – 274. Chem. Phys. 2012, 14, 9377 – 9388; c) A. E. Lewandowska,
[182] a) L. Moens, P. Ruiz, B. Delmon, M. Devillers, Appl. Catal. A M. A. Banares, D. F. Khabibulin, O. B. Lapina, J. Phys. Chem. C
2003, 249, 365 – 374; b) L. Moens, P. Ruiz, B. Delmon, M. 2009, 113, 20648 – 20656; d) A. Corma, J. M. Lopez Nieto, N.
Devillers, Appl. Catal. A 1998, 171, 131 – 143; c) E. A. Mame- Paredes, Appl. Catal. A 1993, 104, 161 – 174.
dov, Appl. Catal. A 1994, 116, 49 – 70. [195] a) T. Cotter, B. Frank, W. Zhang, R. Schlçgl, A. Trunschke,
[183] a) R. J. Berger, F. Kapteijn, J. A. Moulijn, G. B. Marin, J. Chem. Mater. 2013, 25, 3124 – 3136; b) G. A. Zenkovets, G. N.
De Wilde, M. Olea, D. Chen, A. Holmen, L. Lietti, E. Tronconi, Kryukova, V. Y. Gavrilov, S. V. Tsybulya, V. A. Anufrienko,
Y. Schuurman, Appl. Catal. A 2008, 342, 3 – 28; b) R. Fushimi, T. A. Larina, D. F. Khabibulin, O. B. Lapina, E. Rçdel, A.
S. O. Shekhtman, A. Gaffney, S. Han, G. S. Yablonsky, J. T. Trunschke, T. Ressler, R. Schlçgl, Mater. Chem. Phys. 2007, 103,
Gleaves, Ind. Eng. Chem. Res. 2005, 44, 6310 – 6319; c) E. P. J. 295 – 304; c) C. Rameshan, H. Lorenz, L. Mayr, S. Penner, D.
Mallens, J. Hoebink, G. B. Marin, J. Catal. 1996, 160, 222 – 234; Zemlyanov, R. Arrigo, M. Hvecker, A. Knop-Gericke, R.
d) O. V. Buyevskaya, M. Rothaemel, H. W. Zanthoff, M. Schlçgl, B. Klçtzer, J. Catal. 2012, 295, 186 – 194; d) B. Johnson,
Baerns, J. Catal. 1994, 146, 346 – 357; e) P. A. Agaskar, L. F. Girgsdies, G. Weinberg, D. Rosenthal, A. Knop-Gericke, R.
DeCaul, R. K. Grasselli, Catal. Lett. 1994, 23, 339 – 351. Schloegl, T. Reier, P. Strasser, J. Phys. Chem. C 2013, 117,
[184] a) Z. W. Zhu, D. R. Butcher, B. H. Mao, Z. Liu, M. Salmeron, 25443 – 25450.
G. A. Somorjai, J. Phys. Chem. C 2013, 117, 2799 – 2804; [196] D. Teschner, Z. Rvay, J. Borsodi, M. Hvecker, A. Knop-
b) D. R. Butcher, Z. W. Zhu, B. H. Mao, H. L. Wang, Z. Liu, M. Gericke, R. Schlçgl, D. Milroy, S. D. Jackson, D. Torres, P.
Salmeron, G. A. Somorjai, Chem. Commun. 2013, 49, 6903 – Sautet, Angew. Chem. Int. Ed. 2008, 47, 9274 – 9278; Angew.
6905; c) G. A. Somorjai, K. S. Hwang, J. S. Parker, Top. Catal. Chem. 2008, 120, 9414 – 9418.
2003, 26, 87 – 99. [197] a) R. Arrigo, S. Wrabetz, M. E. Schuster, D. Wang, A. Villa, D.
[185] G. Rupprechter, T. Dellwig, H. Unterhalt, H. J. Freund, J. Phys. Rosenthal, R. Schlçgl, D. S. Su, Phys. Chem. Chem. Phys. 2012,
Chem. B 2001, 105, 3797 – 3802. 14, 10523 – 10532; b) R. Arrigo, M. E. Schuster, S. Wrabetz, F.
[186] M. Hvecker, A. Knop-Gericke, T. Schedel-Niedrig, R. Schlçgl, Girgsdies, J.-P. Tessonnier, G. Centi, S. Perathoner, D. S. Su, R.
Angew. Chem. Int. Ed. 1998, 37, 1939 – 1942; Angew. Chem. Schlçgl, ChemSusChem 2012, 5, 577 – 586.
1998, 110, 2049 – 2053. [198] a) H. Soerijanto, C. Rçdel, U. Wild, M. Lerch, R. Schomcker,
[187] a) C. Papp, H. P. Steinruck, Surf. Sci. Rep. 2013, 68, 446 – 487; R. Schlçgl, T. Ressler, J. Catal. 2007, 250, 19 – 24; b) H.
b) M. Salmeron, R. Schlçgl, Surf. Sci. Rep. 2008, 63, 169 – 199; Soerijanto, C. Rçdel, U. Wild, M. Lerch, R. Schomcker, R.
c) H. Bluhm, M. Hvecker, A. Knop-Gericke, M. Kiskinova, R. Schlçgl, T. Ressler, Z. Anorg. Allg. Chem. 2006, 632, 2157; c) B.
Schlçgl, M. Salmeron, MRS Bull. 2007, 32, 1022 – 1030; d) H. Frank, T. P. Cotter, M. E. Schuster, R. Schlçgl, A. Trunschke,
Bluhm, M. Hvecker, A. Knop-Gericke, D. Teschner, E. Chem. Eur. J. 2013, 19, 16938 – 16945; d) W. Zheng, T. P. Cotter,
Kleimenov, V. I. Bukhtiyarov, D. F. Ogletree, M. Salmeron, R. P. Kaghazchi, T. Jacob, B. Frank, K. Schlichte, W. Zhang, D. S.
Schlçgl, BESSY Annual Report 2002 2003, 229 – 232; e) A. Su, F. Schth, R. Schlçgl, J. Am. Chem. Soc. 2013, 135, 3458 –
Knop-Gericke, E. Kleimenov, M. Hvecker, R. Blume, D. 3464; e) S. A. Steiner III, T. F. Baumann, B. C. Bayer, R.
Teschner, S. Zafeiratos, R. Schlçgl, V. I. Bukhtiyarov, V. V. Blume, M. A. Worsley, W. A. MoberlyChan, E. L. Shaw, R.
Kaichev, I. P. Prosvirin, A. I. Nizovskii, H. Bluhm, A. Barinov, Schlçgl, A. J. Hart, S. Hofmann, B. L. Wardle, J. Am. Chem.
P. Dudin, M. Kiskinova, Adv. Catal. 2009, 52, 213 – 272. Soc. 2009, 131, 12144 – 12154.
[188] J. Pantfçrder, S. Pçllmann, J. F. Zhu, D. Borgmann, R. [199] a) J. A. Lopez-Sanchez, J. Bartley, A. Burrows, C. J. Kiely, M.
Denecke, H. P. Steinrck, Rev. Scientific Instrum. 2005, 76, Hvecker, R. Schlçgl, J. C. Volta, M. Poliakoff, G. J. Hutchings,
014102. New J. Chem. 2002, 26, 1811 – 1816; b) G. Landi, L. Lisi, J. C.
[189] a) M. Wilde, K. Fukutani, M. Naschitzki, H. J. Freund, Phys. Volta, J. Mol. Catal. A 2004, 222, 175 – 181; c) G. Landi, L. Lisi,
Rev. B 2008, 77, 113412; b) N. A. Khan, S. Shaikhutdinov, H. J. J. C. Volta, Chem. Commun. 2003, 492 – 493.
Freund, Catal. Lett. 2006, 108, 159 – 164; c) M. Morkel, G. [200] a) F. Ivars, B. Solsona, E. Rodr guez-Castell n, J. M. L pez
Rupprechter, H. J. Freund, Surf. Sci. 2005, 588, L209 – L219. Nieto, J. Catal. 2009, 262, 35 – 43; b) P. Botella, A. Dejoz, M. C.
[190] a) W. D. Williams, M. Shekhar, W. S. Lee, V. Kispersky, W. N. Abello, M. I. Vzquez, L. Arrffla, J. M. Lopez Nieto, Catal.
Delgass, F. H. Ribeiro, S. M. Kim, E. A. Stach, J. T. Miller, L. F. Today 2009, 142, 272 – 277; c) J. M. Lopez Nieto, J. M. Oliver, P.
Allard, J. Am. Chem. Soc. 2010, 132, 14018 – 14020; b) M. Botella, Abstracts of Papers, 226th ACS National Meeting, New
Kotobuki, R. Leppelt, D. A. Hansgen, D. Widmann, R. J. York, NY, United States, September 7 – 11, 2003, COLL-275;
Behm, J. Catal. 2009, 264, 67 – 76; c) G. J. Hutchings, J. Mater. d) A. Corma, J. M. Lopez Nieto, N. Paredes, J. Catal. 1993, 144,
Chem. 2009, 19, 1222 – 1235; d) A. A. Herzing, C. J. Kiely, A. F. 425 – 438.
Carley, P. Landon, G. J. Hutchings, Science 2008, 321, 1331 – [201] F. Girgsdies, R. Schlçgl, A. Trunschke, Catal. Commun. 2012,
1335. 18, 60 – 62.
[191] a) M. V. Ganduglia-Pirovano, C. Popa, J. Sauer, H. Abbott, A. [202] J. B. Wagner, S. B. Abd Hamid, D. Othman, O. Timpe, A.
Uhl, M. Baron, D. Stacchiola, O. Bondarchuk, S. Shaikhutdi- Knop-Gericke, D. Niemeyer, D. S. Su, R. Schlçgl, J. Catal. 2004,
nov, H.-J. Freund, J. Am. Chem. Soc. 2010, 132, 2345 – 2349; 225, 78 – 85.
b) M. Baron, H. Abbott, O. Bondarchuk, D. Stacchiola, A. Uhl, [203] B. Kubias, M. J. G. Fait, R. Schlçgl in Handbook of Heteroge-
S. Shaikhutdinov, H.-J. Freund, C. Popa, M. V. Ganduglia- neous Catalysis, Vol. 1 (Eds.: G. Ertl, H. Knçzinger, F. Schth, J.
Pirovano, J. Sauer, Angew. Chem. Int. Ed. 2009, 48, 8006 – 8009; Weitkamp), Wiley-VCH, Weinheim, 2008, pp. 571 – 583.

3514 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

[204] a) C. Heine, M. Hvecker, M. Sanchez-Sanchez, A. Trunschke, S. B. Abd Hamid, B. Kniep, R. Fischer, W. Busser, M. Muhler,
R. Schlçgl, M. Eichelbaum, J. Phys. Chem. C 2013, 117, 26988 – R. Schlçgl, Appl. Catal. A 2011, 392, 93 – 102.
26997; b) M. Eichelbaum, M. Hvecker, C. Heine, A. Karpov, [212] A. Von Oertzen, A. S. Mikhailov, H. Rotermund, G. Ertl, J.
C.-K. Dobner, F. Rosowski, A. Trunschke, R. Schlçgl, Angew. Phys. Chem. B 1998, 102, 4966 – 4981.
Chem. Int. Ed. 2012, 51, 6246 – 6250; Angew. Chem. 2012, 124, [213] T. Kandemir, M. E. Schuster, A. Senyshyn, M. Behrens, R.
6350 – 6354. Schlçgl, Angew. Chem. Int. Ed. 2013, 52, 12723 – 12726; Angew.
[205] a) J. Wintterlin, J. Trost, S. Renisch, R. Schuster, T. Zambelli, G. Chem. 2013, 125, 12955 – 12959.
Ertl, Surf. Sci. 1997, 394, 159 – 169; b) G. Ertl, Faraday Discuss. [214] A. J. Nagy, G. Mestl, R. Schlçgl, J. Catal. 1999, 188, 58 – 68.
2002, 121, 1 – 15; c) C. Sachs, M. Hildebrand, S. Volkening, J. [215] H. Werner, D. Herein, G. Schulz, U. Wild, R. Schlçgl, Catal.
Wintterlin, G. Ertl, Science 2001, 293, 1635 – 1638. Lett. 1997, 49, 109 – 119.
[206] J. D. Burrington, C. T. Kartisek, R. K. Grasselli, J. Catal. 1982, [216] T. C. R. Rocha, A. Oesterreich, D. V. Demidov, M. Hvecker,
75, 225 – 232. S. Zafeiratos, G. Weinberg, V. I. Bukhtiyarov, A. Knop-
[207] a) S. F. Ye, F. Neese, Curr. Opin. Chem. Biol. 2009, 13, 89 – 98; Gericke, R. Schlçgl, Phys. Chem. Chem. Phys. 2012, 14,
b) M. J. Henson, P. Mukherjee, D. E. Root, T. D. P. Stack, E. I. 4554 – 4564.
Solomon, J. Am. Chem. Soc. 1999, 121, 10332 – 10345. [217] S. Piccinin, S. Zafeiratos, C. Stampfl, T. W. Hansen, M.
[208] a) S. Hofmann, R. Blume, C. T. Wirth, M. Cantoro, R. Sharma, Hvecker, D. Teschner, V. I. Bukhtiyarov, F. Girgsdies, A.
C. Ducati, M. Hvecker, S. Zafeiratos, P. Schnçrch, A. Knop-Gericke, R. Schlçgl, M. Scheffler, Phys. Rev. Lett. 2010,
Oestereich, D. Teschner, M. Albrecht, A. Knop-Gericke, R. 104, 035503.
Schlçgl, J. Robertson, J. Phys. Chem. C 2009, 113, 1648 – 1656; [218] D. Teschner, J. Borsodi, A. Wootsch, Z. Rvay, M. Hvecker, A.
b) B.-S. Kim, E.-Y. Kim, H.-S. Jeon, H.-I. Lee, J.-C. Lee, Mater. Knop-Gericke, S. D. Jackson, R. Schlçgl, Science 2008, 320, 86 –
Trans. 2008, 49, 2147 – 2152; c) R. Kiebach, N. Pienack, W. 89.
Bensch, J.-D. Grunwaldt, A. Michailovski, A. Baiker, T. Fox, Y. [219] Z. Pal, R. Schlçgl, Surf. Sci. 2009, 603, 1793 – 1801.
Zhou, G. R. Patzke, Chem. Mater. 2008, 20, 3022 – 3033; [220] R. Blume, M. Hvecker, S. Zafeiratos, D. Teschner, E. Vass, P.
d) C. X. Wang, G. W. Yang, Mater. Sci. Eng. R 2005, 49, 157 – Schnçrch, A. Knop-Gericke, R. Schlçgl, S. Lizzit, P. Dudin, A.
202; e) M. K. van der Lee, A. J. van Dillen, J. H. Bitter, K. P. Barinov, M. Kiskinova, Phys. Chem. Chem. Phys. 2007, 9, 3648 –
de Jong, J. Am. Chem. Soc. 2005, 127, 13573 – 13582; f) X. S. 3657.
Fang, C. H. Ye, L. D. Zhang, J. X. Zhang, J. W. Zhao, P. Yan, [221] R. Blume, M. Hvecker, S. Zafeiratos, D. Teschner, E.
Small 2005, 1, 422 – 428; g) M. M. Gnter, B. Bems, R. Schlçgl, Kleimenov, A. Knop-Gericke, R. Schlçgl, A. Barinov, P.
T. Ressler, J. Synchrotron Radiat. 2001, 8, 619 – 621; h) J. C. Dudin, M. Kiskinova, J. Catal. 2006, 239, 354 – 361.
Charlier, S. Iijima, Carbon Nanotubes 2001, 80, 55 – 80; i) A. A. [222] E. Kleimenov, H. Bluhm, M. Hvecker, A. Knop-Gericke, A.
Schmidt, R. Anton, Surf. Sci. 1995, 322, 307 – 324. Pestryakov, D. Teschner, J. A. Lopez-Sanches, J. K. Bartley, G.
[209] a) M. Behrens, S. Kißner, F. Girgsdies, I. Kasatkin, F. Her- Hutchings, R. Schlçgl, Surf. Sci. 2005, 575, 181 – 188.
merschmidt, K. Mette, H. Ruland, M. Muhler, R. Schlçgl, [223] M. Hvecker, N. Pinna, K. Weiß, H. Sack-Kongehl, R. E.
Chem. Commun. 2011, 47, 1701 – 1703; b) M. Becker, W. Xia, Jentoft, D. Wang, M. Swoboda, U. Wild, M. Niederberger, J.
J.-P. Tessonnier, L. Yao, R. Schlçgl, M. Muhler, Carbon 2011, Urban, D. S. Su, R. Schlçgl, J. Catal. 2005, 236, 221 – 232.
49, 5253 – 5264; c) M. Schur, B. Bems, A. Dassenoy, I. [224] A. Celaya Sanfiz, T. W. Hansen, D. Teschner, P. Schnçrch, F.
Kassatkine, J. Urban, H. Wilmer, O. Hinrichsen, M. Muhler, Girgsdies, A. Trunschke, R. Schlçgl, M. H. Looi, S. B. Abd Ha-
R. Schlçgl, Angew. Chem. Int. Ed. 2003, 42, 3815 – 3817; Angew. mid, J. Phys. Chem. C 2010, 114, 1912 – 1921.
Chem. 2003, 115, 3945 – 3947; d) H. Schubert, U. Tegtmeyer, D. [225] M. Armbrster, M. Behrens, F. Cinquini, K. Fçttinger, Y. Grin,
Herein, X. Bao, M. Muhler, R. Schlçgl, Catal. Lett. 1995, 33, A. Haghofer, B. Klçtzer, A. Knop-Gericke, H. Lorenz, A. Ota,
305 – 319; e) Z. Pal, X. L. Xu, J. Pal-Lukcs, W. Vogel, M. S. Penner, J. Prinz, C. Rameshan, Z. Rvay, D. Rosenthal, G.
Muhler, R. Schlçgl, J. Catal. 1995, 152, 252 – 263; f) M. Rupprechter, P. Sautet, R. Schlçgl, L. Shao, L. Szentimikl si, D.
Sanchez Sanchez, F. Girgsdies, M. Jastak, P. Kube, R. Schlçgl, Teschner, D. Torres, R. Wagner, R. Widmer, G. Wowsnick,
A. Trunschke, Angew. Chem. Int. Ed. 2012, 51, 7194 – 7197; ChemCatChem 2012, 4, 1048 – 1063.
Angew. Chem. 2012, 124, 7306 – 7309; g) R. Glaum, C. Welker- [226] a) B. Halevi, E. J. Peterson, A. Roy, A. DeLariva, E. Jeroro, F.
Niewoudt, C.-K. Dobner, M. Eichelbaum, F. Gruchow, C. Gao, Y. Wang, J. M. Vohs, B. Kiefer, E. Kunkes, M. Hvecker,
Heine, A. Karpov, R. Kniep, F. Rosowski, R. Schlçgl, S. A. M. Behrens, R. Schlçgl, A. K. Datye, J. Catal. 2012, 291, 44 – 54;
Schunk, S. Titlbach, A. Trunschke, Chem. Ing. Tech. 2012, 84, b) A. Ota, M. Armbrster, M. Behrens, D. Rosenthal, M.
1766; h) M. Behrens, D. Brennecke, F. Girgsdies, S. Kißner, A. Friedrich, I. Kasatkin, F. Girgsdies, W. Zhang, R. Wagner, R.
Trunschke, N. Nasrudin, S. Zakaria, N. F. Idris, S. B. Abd Ha- Schlçgl, J. Phys. Chem. C 2011, 115, 1368 – 1374; c) M.
mid, B. Kniep, R. Fischer, W. Busser, M. Muhler, R. Schlçgl, Armbrster, K. Kovnir, M. Behrens, D. Teschner, Y. Grin, R.
Appl. Catal. A 2011, 392, 93 – 102; i) E. Rçdel, O. Timpe, A. Schlçgl, J. Am. Chem. Soc. 2010, 132, 14745 – 14747; d) M.
Trunschke, G. A. Zenkovets, G. N. Kryukova, R. Schlçgl, T. Armbrster, K. Kovnir, M. Friedrich, D. Teschner, G. Wows-
Ressler, Catal. Today 2007, 126, 318 – 327. nick, M. Hahne, P. Gille, L. Szentmikl si, M. Feuerbacher, M.
[210] a) B. Bems, M. Schur, A. Dassenoy, H. Junkes, D. Herein, R. Heggen, F. Giurgsdies, D. Rosenthal, R. Schlçgl, Y. Grin, Nat.
Schlçgl, Chem. Eur. J. 2003, 9, 2039 – 2052; b) Y. V. Kolenko, Mater. 2012, 11, 690 – 693; e) L. Shao, W. Zhang, M. Armbrs-
K. Amakawa, R. Naumann dAlnoncourt, F. Girgsdies, G. ter, D. Teschner, F. Girgsdies, B. Zhang, O. Timpe, M. Friedrich,
Weinberg, R. Schlçgl, A. Trunschke, ChemCatChem 2012, 4, R. Schlçgl, D. S. Su, Angew. Chem. Int. Ed. 2011, 50, 10231 –
495 – 503; c) N. Hamilton, T. Wolfram, G. Tzolova-Mller, M. 10235; Angew. Chem. 2011, 123, 10414 – 10418.
Hvecker, J. Krçhnert, C. Carrero, R. Schomcker, A. [227] D. Teschner, G. Novell-Leruth, R. Farra, A. Knop-Gericke, R.
Trunschke, R. Schlçgl, Catal. Sci. Technol. 2012, 2, 1346 – 1359. Schlçgl, L. Szentmiklosi, M. G. Hevia, H. Soerijanto, R.
[211] a) F. Schth, Chem. Ing. Tech. 2005, 77, 1399 – 1416; b) Y. V. Schomacker, J. Perez-Ramirez, N. Lopez, Nat. Chem. 2012, 4,
Kolenko, W. Zhang, R. N. dAlnoncourt, F. Girgsdies, T. W. 739 – 745.
Hansen, T. Wolfram, R. Schlçgl, A. Trunschke, ChemCatChem [228] a) R. Naumann dAlnoncourt, Y. V. Kolenko, R. Schlçgl, A.
2011, 3, 1597 – 1606; c) M. Behrens, D. Brennecke, F. Girgsdies, Trunschke, Comb. Chem. High Throughput Screening 2012, 15,
S. Kissner, A. Trunschke, N. Nasrudin, S. Zakaria, N. F. Idris, 161 – 169; b) R. Naumann dAlnoncourt, L.-I. Csepei, M.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3515
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Haevecker, F. Girgsdies, M. E. Schuster, R. Schloegl, A. Studies in Surf. Sci. and Catalysis, Vol. 112, Elsevier, Amster-
Trunschke, J. Catal. 2014, 311, 369 – 385. dam, 1997, S. 473 – 480; d) Y. V. Kolenko, W. Zhang, R.
[229] N. D. Spencer, R. C. Schoonmaker, G. A. Somorjai, J. Catal. Naumann dAlnoncourt, F. Girgsdies, T. W. Hansen, T. Wolf-
1982, 74, 129 – 135. ram, R. Schlçgl, A. Trunschke, ChemCatChem 2011, 3, 1597 –
[230] a) H. L. Xin, A. Vojvodic, J. Voss, J. K. Norskov, F. Abild- 15458.
Pedersen, Phys. Rev. B 2014, 89, 115114; b) D. Teschner, U. [249] a) J. C. Vedrine, G. Coudurier, J.-M. M. Millet, Catal. Today
Wild, R. Schlçgl, Z. Pal, J. Phys. Chem. B 2005, 109, 20516 – 1997, 33, 3 – 13; b) J. C. Vedrine, J. M. M. Millet, J.-C. Volta,
20521; c) C. Tsai, K. Chan, J. K. Norskov, F. Abild-Pedersen, J. Catal. Today 1996, 32, 115 – 123.
Phys. Chem. Lett. 2014, 5, 3884 – 3889. [250] a) J. C. Vdrine, Top. Catal. 2002, 21, 97 – 106; b) K. Tanabe,
[231] W. Thiel, Angew. Chem. Int. Ed. 2014, 53, 8605 – 8613; Angew. W. F. Holderich, Appl. Catal. A 1999, 181, 399 – 434; c) F. Eder,
Chem. 2014, 126, 8748 – 8757. J. A. Lercher, Zeolites 1997, 18, 75 – 81; d) H. Hattori, Chem.
[232] F. R. Negreiros, G. Barcaro, L. Sementa, A. Fortunelli, C. R. Rev. 1995, 95, 537 – 558; e) H. Noller, J. A. Lercher, H. Vinek,
Chim. 2014, 17, 625 – 633. Mater. Chem. Phys. 1988, 18, 577 – 593.
[233] E. C. Beret, L. M. Ghiringhelli, M. Scheffler, Faraday Discuss. [251] B. List, Chem. Commun. 2006, 819 – 824.
2011, 152, 153 – 167. [252] a) J. Llorens, M. Pera-Titus, J. Colloid Interface Sci. 2009, 331,
[234] A. M. H. Rasmussen, M. N. Groves, B. Hammer, ACS Catal. 302 – 311; b) E. R. Parnham, R. E. Morris, Acc. Chem. Res.
2014, 4, 1182 – 1188. 2007, 40, 1005 – 1013; c) T. Wakihara, T. Okubo, Chem. Lett.
[235] B. Temel, H. Meskine, K. Reuter, M. Scheffler, H. Metiu, J. 2005, 34, 276 – 281; d) L. Tosheva, V. P. Valtchev, Chem. Mater.
Chem. Phys. 2007, 126, 204711. 2005, 17, 2494 – 2513.
[236] W. Liu, A. Tkatchenko, M. Scheffler, Acc. Chem. Res. 2014, 47, [253] a) M. E. Davis, Nature 2002, 417, 813 – 821; b) A. Corma,
3369 – 3377. Chem. Rev. 1995, 95, 559 – 614; c) P. B. Venuto, Microporous
[237] G. Ertl, M. Huber, N. Thiele, Z. Naturforsch. A 1979, 34, 30 – 39. Mater. 1994, 2, 297 – 411; d) J. S. Beck, J. C. Vartuli, W. J. Roth,
[238] a) Y.-S. Hu, X. Liu, J.-O. Mller, R. Schlçgl, J. Maier, D. S. Su, M. E. Leonowicz, C. T. Kresge, K. D. Schmitt, C. T. W. Chu,
Angew. Chem. Int. Ed. 2009, 48, 210 – 214; Angew. Chem. 2009, D. H. Olson, E. W. Sheppard, S. B. McCullen, J. B. Higgins, J. L.
121, 216 – 220; b) O. Delmer, P. Balaya, L. Kienle, J. Maier, Adv. Schlenker, J. Am. Chem. Soc. 1992, 114, 10834 – 10843; e) D. W.
Mater. 2008, 20, 501; c) J. Jamnik, J. Maier, Phys. Chem. Chem. Breck, W. G. Eversole, R. M. Milton, T. B. Reed, T. L. Thomas,
Phys. 2003, 5, 5215 – 5220. J. Am. Chem. Soc. 1956, 78, 5963 – 5971.
[239] a) J. Maier, Phys. Chem. Chem. Phys. 2009, 11, 3011 – 3022; b) J. [254] a) L. Lichtenstein, C. Buchner, B. Yang, S. Shaikhutdinov, M.
Maier, Solid State Ionics 2002, 148, 367 – 374; c) J. Maier, Chem. Heyde, M. Sierka, R. Wlodarczyk, J. Sauer, H. J. Freund,
Eur. J. 2001, 7, 4762 – 4770. Angew. Chem. Int. Ed. 2012, 51, 404 – 407; Angew. Chem. 2012,
[240] a) J.-M. Herrmann, Catal. Today 2006, 112, 73 – 77; b) M. Anpo, 124, 416 – 420; b) J. A. Boscoboinik, X. Yu, B. Yang, F. D.
M. Che, B. C. G. Werner O. Haag, K. Helmut in Advances in Fischer, R. Wlodarczyk, M. Sierka, S. Shaikhutdinov, J. Sauer,
Catalysis, Vol. 44, Academic Press, 1999, S. 119 – 257. H. J. Freund, Angew. Chem. Int. Ed. 2012, 51, 6005 – 6008;
[241] a) K. Y. Yeh, M. J. Janik, J. Comput. Chem. 2011, 32, 3399 – Angew. Chem. 2012, 124, 6107 – 6111.
3408; b) M. T. M. Koper, J. Electroanal. Chem. 2005, 574, 375 – [255] J. V. Smith, Chem. Rev. 1988, 88, 149 – 182.
386. [256] a) A. Corma, H. Garcia, F. X. L. I. Llabres i Xamena, Chem.
[242] H. L. Chen, H. M. Lee, S. H. Chen, Y. Chao, M. B. Chang, Rev. 2010, 110, 4606 – 4655; b) J. Lee, O. K. Farha, J. Roberts,
Appl. Catal. B 2008, 85, 1 – 9. K. A. Scheidt, S. T. Nguyen, J. T. Hupp, Chem. Soc. Rev. 2009,
[243] a) I. R. Subbotina, V. B. Kazansky, F. C. Jentoft, R. Schlçgl, 38, 1450 – 1459; c) S. L. James, Chem. Soc. Rev. 2003, 32, 276 –
Stud. Surf. Sci. Catal. 2008, 174, 849 – 852; b) V. B. Kazansky, 288; d) A. K. Cheetham, G. Ferey, T. Loiseau, Angew. Chem.
I. R. Subbotina, F. C. Jentoft, R. Schlçgl, J. Phys. Chem. B 2006, Int. Ed. 1999, 38, 3268 – 3292; Angew. Chem. 1999, 111, 3466 –
110, 17468 – 17477. 3492.
[244] a) C. Galeano, J. C. Meier, V. Peinecke, H. Bongard, I. [257] a) H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, P.
Katsounaros, A. A. Topalov, A. Lu, K. J. J. Mayrhofer, F. Strasser, ChemCatChem 2010, 2, 724 – 761; b) C. Herwig, C.
Schueth, J. Am. Chem. Soc. 2012, 134, 20457 – 20465; b) J. H. Limberg, Inorg. Chem. 2008, 47, 2937 – 2939; c) C. Limberg,
Bitter, J. Mater. Chem. 2010, 20, 7312 – 7321; c) S. Valange, R. Angew. Chem. Int. Ed. 2003, 42, 5932 – 5954; Angew. Chem.
Palacio, A. Charmot, J. Barrault, A. Louati, Z. Gabelica, J. Mol. 2003, 115, 6112 – 6136.
Catal. A 2009, 305, 24 – 33; d) J.-P. Tessonnier, O. Ersen, G. [258] A. Indra, P. W. Menezes, I. Zaharieva, E. Baktash, J. Pfrommer,
Weinberg, C. Pham-Huu, D. S. Su, R. Schlçgl, ACS Nano 2009, M. Schwarze, H. Dau, M. Driess, Angew. Chem. Int. Ed. 2013,
3, 2081 – 2089; e) G. Mills, M. S. Gordon, H. Metiu, J. Chem. 52, 13206 – 13210; Angew. Chem. 2013, 125, 13447 – 13451.
Phys. 2003, 118, 4198 – 4205; f) X. L. Pan, X. H. Bao, Acc. [259] a) J. Yu, M. Scheffler, H. Metiu, J. Phys. Chem. C 2013, 117,
Chem. Res. 2011, 44, 553 – 562; g) P. Serp, E. Castillejos, 18475 – 18483; b) L. Jiang, T. Wende, P. Claes, S. Bhattacharyya,
ChemCatChem 2010, 2, 41 – 47; h) X. L. Pan, Z. L. Fan, W. M. Sierka, G. Meijer, P. Lievens, J. Sauer, K. R. Asmis, J. Phys.
Chen, Y. J. Ding, H. Y. Luo, X. H. Bao, Nat. Mater. 2007, 6, Chem. A 2011, 115, 11187 – 11192; c) C. S. Guo, K. Hermann,
507 – 511. M. Havecker, J. P. Thielemann, P. Kube, L. J. Gregoriades, A.
[245] L. Cheng, C. R. Yin, F. Mehmood, B. Liu, J. Greeley, S. Lee, B. Trunschke, J. Sauer, R. Schlçgl, J. Phys. Chem. C 2011, 115,
Lee, S. Seifert, R. E. Winans, D. Teschner, R. Schlçgl, S. Vajda, 15449 – 15458; d) T. K. Todorova, J. Doebler, M. Sierka, J.
L. A. Curtiss, ACS Catal. 2014, 4, 32 – 39. Sauer, J. Phys. Chem. C 2009, 113, 8336 – 8342; e) A. Hofmann,
[246] J. C. Vdrine, Appl. Catal. A 2014, 474, 40 – 50. M. V. Ganduglia-Pirovano, J. Sauer, J. Phys. Chem. C 2009, 113,
[247] W. Kaim, B. Schwederski, Coord. Chem. Rev. 2010, 254, 1580 – 18191 – 18203.
1588. [260] Y. Romanyshyn, S. Guimond, H. Kuhlenbeck, S. Kaya, R.
[248] a) G. Y. Popova, T. V. Andrushkevich, L. S. Dovlitova, G. A. Blum, H. Niehus, S. Shaikhutdinov, V. Simic-Milosevic, N.
Aleshina, Y. A. Chesalov, A. V. Ishenko, E. V. Ishenko, L. M. Nilius, H. J. Freund, M. Ganduglia-Pirovano, R. Fortrie, J.
Plyasova, V. V. Malakhov, M. I. Khramov, Appl. Catal. A 2009, Dçbler, J. Sauer, Top. Catal. 2008, 50, 106 – 115.
353, 249 – 257; b) W. A. Goddard III, K. Chenoweth, S. Pudar, [261] K. Kwapien, M. Sierka, J. Doebler, J. Sauer, M. Haertelt, A.
A. C. T. Duin, M.-J. Cheng, Top. Catal. 2008, 50, 2 – 18; c) T. Fielicke, G. Meijer, Angew. Chem. Int. Ed. 2011, 50, 1716 –
Ushikubo, K. Oshima, A. Kayou, M. Hatano, X. Can Li, Qin in 1719; Angew. Chem. 2011, 123, 1754 – 1757.

3516 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

[262] a) M. Haertelt, A. Fielicke, G. Meijer, K. Kwapien, M. Sierka, J. Birkner, L. Ruppel, C. Wçll, J. Qian, C. Liang, G. Marginean,
Sauer, Phys. Chem. Chem. Phys. 2012, 14, 2849 – 2856; b) D. K. W. Brandl, M. Muhler, Chem. Mater. 2005, 17, 5737 – 5742;
Bçhme, H. Schwarz, Angew. Chem. Int. Ed. 2005, 44, 2336 – d) M. Schur, B. Bems, A. Dassenoy, I. Kassatkine, J. Urban, H.
2354; Angew. Chem. 2005, 117, 2388 – 2406; c) D. Schrçder, H. Wilmes, O. Hinrichsen, M. Muhler, R. Schlçgl, Angew. Chem.
Schwarz, Angew. Chem. Int. Ed. Engl. 1995, 34, 1973 – 1995; Int. Ed. 2003, 42, 3815 – 3817; Angew. Chem. 2003, 115, 3945 –
Angew. Chem. 1995, 107, 2126 – 2150; d) K. Eller, H. Schwarz, 3947.
Chem. Rev. 1991, 91, 1121 – 1177. [271] a) R. Schlçgl in Handbook of Heterogeneous Catalysis, Vol. 5
[263] a) J. M. Thomas, Chem. Eur. J. 1997, 3, 1557 – 1562; b) S. (Eds.: G. Ertl, H. Knçzinger, F. Schth, J. Weitkamp), Wiley-
Gnther, L. Zhou, M. Hvecker, A. Knop-Gericke, E. Klei- VCH, Weinheim, 2008, pp. 2501 – 2575; b) R. Schlçgl, Angew.
menov, R. Schlçgl, R. Imbihl, J. Chem. Phys. 2006, 125, 114709; Chem. Int. Ed. 2003, 42, 2004 – 2008; Angew. Chem. 2003, 115,
c) V. C. Holmberg, M. G. Panthani, B. A. Korgel, Science 2009, 2050 – 2055.
326, 405 – 407; d) D. Vogel, C. Spiel, Y. Suchorski, A. Urich, R. [272] A. Vojvodic, A. J. Medford, F. Studt, F. Abild-Pedersen, T. S.
Schlçgl, G. Rupprechter, Surf. Sci. 2011, 605, 1999 – 2005. Khan, T. Bligaard, J. K. Norskov, Chem. Phys. Lett. 2014, 598,
[264] a) M. W. E. van den Berg, S. Polarz, O. P. Tkachenko, K. 108 – 112.
Kaehler, M. Muhler, W. Gruenert, Catal. Lett. 2009, 128, 49 – [273] a) F. Schth, L. Baumes, F. Clerc, D. Demuth, D. Farrusseng, J.
56; b) H. Over, M. Muhler, Prog. Surf. Sci. 2003, 72, 3 – 17; c) L. Llamas-Galilea, C. Klanner, J. Klein, A. Martinez-Joaristi, J.
Cheng, C. Yin, F. Mehmood, B. Liu, J. Greeley, S. Lee, B. Lee, S. Procelewska, M. Saupe, S. Schunk, M. Schwickardi, W.
Seifert, R. E. Winans, D. Teschner, R. Schlçgl, S. Vajda, L. A. Strehlau, T. Zech, Catal. Today 2006, 117, 284 – 290; b) K.
Curtiss, ACS Catal. 2014, 4, 32 – 39; d) D. Zemlyanov, B. Jhnisch, V. Hessel, H. Lowe, M. Baerns, Angew. Chem. Int. Ed.
Klçtzer, H. Gabasch, A. Smeltz, F. H. Ribeiro, S. Zafeiratos, D. 2004, 43, 406 – 446; Angew. Chem. 2004, 116, 410 – 451.
Teschner, P. Schnçrch, E. Vass, M. Hvecker, A. Knop-Gericke, [274] U. Zavyalova, M. Holena, R. Schlçgl, M. Baerns, Chem-
R. Schlçgl, Top. Catal. 2013, 56, 885 – 895; e) D. Vogel, C. Spiel, CatChem 2011, 3, 1935 – 1947.
M. Schmid, M. Stçger-Pollach, R. Schlçgl, Y. Suchorski, G. [275] D. Widmann, R. J. Behm, Acc. Chem. Res. 2014, 47, 740 – 749.
Rupprechter, J. Phys. Chem. C 2013, 117, 12054 – 12060; f) L. [276] J. M. Thomas, Faraday Discuss. 1996, 105, 1 – 31.
Shao, B. Zhang, W. Zhang, S. Y. Hong, R. Schlçgl, D. S. Su, [277] a) J. K. Nørskov, P. Stoltze, Surf. Sci. 1987, 189/190, 91 – 105;
Angew. Chem. Int. Ed. 2013, 52, 2114 – 2117; Angew. Chem.
b) S. Dahl, A. Logadottir, C. J. H. Jacobsen, J. K. Norskov,
2013, 125, 2168 – 2171; g) R. Weatherup, B. C. Bayer, R. Blume,
Appl. Catal. A 2001, 222, 19 – 29; c) S. Piccinin, S. Zafeiratos, C.
C. Baehtz, P. R. Kidambi, M. Fouquet, C. T. Wirth, R. Schlçgl,
Stampfl, T. W. Hansen, M. Havecker, D. Teschner, V. I.
S. Hofmann, ChemPhysChem 2012, 13, 2544 – 2549.
Bukhtiyarov, F. Girgsdies, A. Knop-Gericke, R. Schlçgl, M.
[265] a) A. Tompos, J. L. Margitfalvi, E. G. Szabo, L. Vegvari, Top.
Scheffler, Phys. Rev. Lett. 2010, 104, 035503.
Catal. 2010, 53, 108 – 115; b) M. Sadakane, W. Ueda, Shokubai
[278] a) T. F. Jaramillo, K. P. Jorgensen, J. Bonde, J. H. Nielsen, S.
2007, 49, 542 – 547; c) M. Neurock, J. Catal. 2003, 216, 73 – 88;
Horch, I. Chorkendorff, Science 2007, 317, 100 – 102; b) K.
d) G. J. A. A. Soler-Illia, et al., Chem. Rev. 2002, 102, 4093 –
Reuter, C. Stampfl, M. V. Ganduglia-Pirovano, M. Scheffler,
4138; e) C. J. H. Jacobsen, S. Dahl, B. S. Clausen, S. Bahn, A.
Chem. Phys. Lett. 2002, 352, 311 – 317; c) V. Stamenkovic, B. S.
Logadottir, J. K. Norskov, J. Am. Chem. Soc. 2001, 123, 8404 –
Mun, K. J. J. Mayrhofer, P. N. Ross, N. M. Markovic, J. Ross-
8405; f) M. A. Carreon, V. V. Guliants, Chem. Commun. 2001,
meisl, J. Greeley, J. K. Norskov, Angew. Chem. Int. Ed. 2006, 45,
1438 – 1439; g) J. M. Thomas, Angew. Chem. Int. Ed. 1999, 38,
2897 – 2901; Angew. Chem. 2006, 118, 2963 – 2967; d) T. H. Rod,
3588 – 3628; Angew. Chem. 1999, 111, 3800 – 3843; h) E. Iglesia,
A. Logadottir, J. K. Norskov, J. Chem. Phys. 2000, 112, 5343 –
Appl. Catal. A 1997, 161, 59 – 78; i) O. M. Yaghi, G. M. Li, H. L.
Li, Nature 1995, 378, 703 – 706; j) R. Schlçgl, Angew. Chem. Int. 5347; e) J. Greeley, J. K. Norskov, Surf. Sci. 2007, 601, 1590 –
Ed. Engl. 1994, 33, 311 – 312; Angew. Chem. 1994, 106, 319 – 1598; f) J. Greeley, T. F. Jaramillo, J. Bonde, I. B. Chorkendorff,
320; k) J. L. G. Fierro, Catal. Lett. 1993, 22, 67 – 91. J. K. Norskov, Nat. Mater. 2006, 5, 909 – 913.
[266] G. M. Schwab, Handbuch der Katalyse, Springer, Wien, 1941. [279] R. K. Grasselli, Top. Catal. 2002, 21, 79 – 88.
[267] a) C. E. Chan Thaw, F. Garin, R. Schlçgl, F. C. Jentoft, [280] a) F. Cavani, J. H. Teles, ChemSusChem 2009, 2, 508 – 534; b) F.
unpublished results; b) Z. Pal, A. Wootsch, R. Schlçgl, U. Cavani, N. Ballarini, A. Cericola, Catal. Today 2007, 127, 113 –
Wild, Appl. Catal. A 2005, 282, 135 – 145; c) R. Quinn, T. A. 131.
Dahl, B. A. Toseland, Appl. Catal. A 2004, 272, 61 – 68; d) C. [281] a) W. D. Pyrz, D. A. Blom, T. Vogt, D. J. Buttrey, Angew. Chem.
Glasson, C. Geantet, M. Lacroix, F. Labruyere, P. Dufresne, J. Int. Ed. 2008, 47, 2788 – 2791; Angew. Chem. 2008, 120, 2830 –
Catal. 2002, 212, 76 – 85; e) B. S. Klose, F. C. Jentoft, P. Joshi, A. 2833; b) P. DeSanto, Jr., D. J. Buttrey, R. K. Grasselli, C. G.
Trunschke, R. Schlçgl, I. R. Subbotina, V. B. Kazansky, Catal. Lugmair, A. F. Volpe, Jr., B. H. Toby, T. Vogt, Z. Kristallogr.
Today 2006, 116, 121 – 131; f) B. S. Klose, F. C. Jentoft, R. 2004, 219, 152 – 165.
Schlçgl, J. Catal. 2005, 233, 68 – 80; g) A. Barański, A. Kotarba, [282] a) W. Zhang, A. Trunschke, R. Schlçgl, D. S. Su, Angew. Chem.
J. M. Łagan, A. Pattek-Janczyk, E. Pyrczak, A. Reizer, Appl. Int. Ed. 2010, 49, 6084 – 6089; Angew. Chem. 2010, 122, 6220 –
Catal. A 1994, 112, 13 – 36. 6225; b) A. Celaya Sanfiz, T. W. Hansen, D. Teschner, P.
[268] a) Y. Wang, X. Wang, M. Antonietti, Angew. Chem. Int. Ed. Schnçrch, F. Girgsdies, A. Trunschke, R. Schlçgl, M. H. Looi,
2012, 51, 68 – 89; Angew. Chem. 2012, 124, 70 – 92; b) C. S. B. A. Hamid, J. Phys. Chem. C 2010, 114, 1912 – 1921.
Giordano, M. Antonietti, Nano Today 2011, 6, 366 – 380; [283] R. Naumann dAlnoncourt, L. I. Csepei, M. Havecker, F.
c) M. M. Titirici, A. Thomas, S. H. Yu, J. O. Muller, M. Girgsdies, M. E. Schuster, R. Schlçgl, A. Trunschke, J. Catal.
Antonietti, Chem. Mater. 2007, 19, 4205 – 4212; d) J. Polleux, 2014, 311, 369 – 385.
N. Pinna, M. Antonietti, C. Hess, U. Wild, R. Schlçgl, M. [284] a) M. Abon, K. E. Bere, A. Tuel, P. Delichere, J. Catal. 1995,
Niederberger, Chem. Eur. J. 2005, 11, 3541 – 3551. 156, 28 – 36; b) G. Centi, Catal. Today 1993, 16, 5 – 26; c) E.
[269] F. Schth, Chem. Mater. 2001, 13, 3184 – 3195. Bordes, Catal. Today 1993, 16, 27 – 38; d) N. H. Batis, H. Batis,
[270] a) S. Kaluza, M. Behrens, N. Schiefenhovel, B. Kniep, R. A. Ghorbel, J. C. Vedrine, J. C. Volta, J. Catal. 1991, 128, 248 –
Fischer, R. Schlçgl, M. Muhler, ChemCatChem 2011, 3, 189 – 263; e) M. G. Willinger, D. S. Su, R. Schlçgl, Phys. Rev. B 2005,
199; b) M. Behrens, A. Furche, I. Kasatkin, A. Trunschke, W. 71, 155118; f) H. Bluhm, M. Hvecker, E. Kleimenov, A. Knop-
Busser, M. Muhler, B. Kniep, R. Fischer, R. Schlçgl, Chem- Gericke, A. Liskowski, R. Schlçgl, D. S. Su, Top. Catal. 2003, 23,
CatChem 2010, 2, 816 – 818; c) W. Xia, D. S. Su, R. Schlçgl, A. 99 – 107; g) M. Hvecker, A. Knop-Gericke, R. W. Mayer, M.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3517
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

Fait, H. Bluhm, R. Schlçgl, J. Electron Spectrosc. Relat. Fottinger, Y. Grin, A. Haghofer, B. Klotzer, A. Knop-Gericke,
Phenom. 2002, 125, 79 – 87. H. Lorenz, A. Ota, S. Penner, J. Prinz, C. Rameshan, Z. Revay,
[285] W. S. Dong, J. K. Bartley, N. F. Dummer, F. Girgsdies, D. S. Su, D. Rosenthal, N. Rupprechter, P. Sautet, R. Schlçgl, L. D. Shao,
R. Schlçgl, J. C. Volta, G. J. Hutchings, J. Mater. Chem. 2005, 15, L. Szentmiklosi, D. Teschner, D. Torres, R. Wagner, R. Widmer,
3214 – 3220. G. Wowsnick, ChemCatChem 2012, 4, 1048 – 1063; e) K.
[286] a) M. Conte, G. Budroni, J. K. Bartley, S. H. Taylor, A. F. Kovnir, M. Armbrster, D. Teschner, T. Venkov, L. Szentmi-
Carley, A. Schmidt, D. M. Murphy, F. Girgsdies, T. Ressler, R. kl si, F. C. Jentoft, A. Knop-Gericke, Y. Grin, R. Schlçgl, Surf.
Schlçgl, G. J. Hutchings, Science 2006, 313, 1270 – 1273; b) M. Sci. 2009, 603, 1784 – 1792; f) K. Kovnir, M. Armbrster, D.
Hvecker, R. W. Mayer, A. Knop-Gericke, H. Bluhm, E. Teschner, T. V. Venkov, F. C. Jentoft, A. Knop-Gericke, Y.
Kleimenov, A. Lisowski, D. S. Su, R. Follath, F. G. Requejo, Grin, R. Schlçgl, Sci. Technol. Adv. Mater. 2007, 8, 420 – 427.
D. F. Ogletree, M. Salmeron, J. A. Lopez-Sanches, J. K. Bartley, [301] a) J. Osswald, K. Kovnir, M. Armbrster, R. Giedigleit, R. E.
G. J. Hutchings, R. Schlçgl, J. Phys. Chem. B 2003, 107, 4587 – Jentoft, U. Wild, Y. Grin, R. Schlçgl, J. Catal. 2008, 258, 219 –
4596. 227; b) J. Osswald, R. Giedigkeit, R. E. Jentoft, M. Armbrster,
[287] a) T. Okuhara, M. Misono, Catal. Today 1993, 16, 61 – 67; b) A. F. Girgsdies, K. Kovnir, T. Ressler, Y. Grin, R. Schlçgl, J. Catal.
Brckner, Top. Catal. 2006, 38, 133 – 139. 2008, 258, 210 – 218.
[288] G. W. Coulston, S. R. Bare, H. Kung, K. Birkeland, G. K. [302] M. Armbrster, K. Kovnir, M. Behrens, D. Teschner, Y. Grin,
Bethke, R. Harlow, N. Herron, P. L. Lee, Science 1997, 275, R. Schlçgl, J. Am. Chem. Soc. 2010, 132, 14745 – 14747.
191 – 193. [303] K. Kovnir, M. Armbrster, D. Teschner, T. Venkov, L.
[289] a) M. Hvecker, M. Cavalleri, R. Herbert, R. Follath, A. Knop- Szentmikl si, F. C. Jentoft, A. Knop-Gericke, Y. Grin, R.
Gericke, C. Hess, K. Hermann, R. Schlçgl, Phys. Status Solidi B Schlçgl, Surf. Sci. 2009, 603, 1784 – 1792.
2009, 246, 1459 – 1469; b) E. Kleimenov, H. Bluhm, M. [304] D. Rosenthal, R. Widmer, R. Wagner, P. Gille, M. Armbruester,
Hvecker, A. Knop-Gericke, A. Pestryakov, D. Teschner, Y. Grin, R. Schlçgl, O. Groening, Langmuir 2012, 28, 6848 –
J. A. Lopez-Sanches, J. K. Bartley, G. Hutchings, R. Schlçgl, 6856.
Surf. Sci. 2005, 575, 181 – 188; c) M. Hvecker, N. Pinna, K. [305] M. Armbrster, K. Kovnir, Y. Grin, R. Schlçgl, in Complex
Weiß, H. Sack-Kongehl, R. E. Jentoft, D. Wang, M. Swoboda, Metallic Alloys. Fundamentals and Applications, Vol. XXIV
U. Wild, M. Niederberger, J. Urban, D. S. Su, R. Schlçgl, J.
(Eds.: J.-M. Dubois, E. Belin-Ferr), Wiley-VCH, Weinheim,
Catal. 2005, 236, 221 – 232.
2010, pp. 395 – 397.
[290] a) S. Ieda, S. Phiyanalinmat, S.-I. Komai, T. Hattori, A.
[306] A. Ota, J. Krçhnert, G. Weinberg, I. Kasatkin, E. L. Kunkes, D.
Satsuma, J. Catal. 2005, 236, 304 – 312; b) V. V. Guliants, S. A.
Ferri, F. Girgsdies, N. Hamilton, M. Armbruester, R. Schlçgl,
Holmes, J. Mol. Catal. A 2001, 175, 227 – 239.
M. Behrens, ACS Catal. 2014, 4, 2048 – 2059.
[291] a) V. V. Guliants, J. B. Benziger, S. Sundaresan, I. E. Wachs,
[307] R. Leary, Z. Saghi, M. Armbrster, G. Wowsnick, R. Schlçgl, J.
J. M. Jehng, J. E. Roberts, Catal. Today 1996, 28, 275 – 295;
Thomas, P. Midgley, J. Phys. Chem. C 2012, 116, 13343 – 13352.
b) V. V. Guliants, J. B. Benziger, S. Sundaresan, N. Yao, I. E.
[308] M. Bauer, R. Schoch, L. Shao, B. Zhang, A. Knop-Gericke, M.
Wachs, Catal. Lett. 1995, 32, 379 – 386.
Willinger, R. Schlçgl, D. Teschner, J. Phys. Chem. C 2012, 116,
[292] F. Cavani, S. Ligi, T. Monti, F. Pierelli, F. Trifir , S. Albonetti, G.
22375 – 22385.
Mazzoni, Catal. Today 2000, 61, 203 – 210.
[309] a) R. Schlçgl, K. Noack, H. Zbinden, A. Reller, Helv. Chim.
[293] H. Molero, B. F. Bartlett, W. T. Tysoe, J. Catal. 1999, 181, 49 –
Acta 1987, 70, 627 – 679; b) J. Rajaram, P. S. Narula, H. P. S.
56.
Chawla, S. Dev, Tetrahedron 1983, 39, 2315 – 2322; c) H.
[294] S. Azad, M. Kaltchev, D. Stacchiola, G. Wu, W. T. Tysoe, J. Phys.
Lindlar, Helv. Chim. Acta 1952, 35, 446 – 450.
Chem. B 2000, 104, 3107 – 3115.
[310] L. Shao, B. S. Zhang, W. Zhang, D. Teschner, F. Girgsdies, R.
[295] a) A. Mastalir, Z. Kiraly, J. Catal. 2003, 220, 372 – 381; b) A.
Schlçgl, D. S. Su, Chem. Eur. J. 2012, 18, 14962 – 14966.
Sarkany, A. H. Weiss, L. Guczi, J. Catal. 1986, 98, 550 – 553;
c) L. H. Dubois, D. G. Castner, G. A. Somorjai, J. Chem. Phys. [311] S. Schauermann, N. Nilius, S. Shaikhutdinov, H.-J. Freund, Acc.
1980, 72, 5234 – 5240; d) G. C. Bond, D. A. Dowden, N. Mack- Chem. Res. 2013, 46, 1673 – 1681.
enzie, Trans. Faraday Soc. 1958, 54, 1537 – 1546. [312] A. Trunschke in Nanostructured Catalysts: Selective Oxidations,
[296] P. A. Sheth, M. Neurock, C. M. Smith, J. Phys. Chem. B 2005, The Royal Society of Chemistry, London, 2011, S. 56 – 95.
109, 12449 – 12466. [313] a) A. Vojvodic, F. Calle-Vallejo, W. Guo, S. Wang, A. Toftelund,
[297] F. Studt, F. Abild-Pedersen, T. Bligaard, R. Z. Sorensen, C. H. F. Studt, J. I. Martinez, J. Shen, I. C. Man, J. Rossmeisl, T.
Christensen, J. K. Norskov, Science 2008, 320, 1320 – 1322. Bligaard, J. K. Nørskov, F. Abild-Pedersen, J. Chem. Phys. 2011,
[298] a) A. A. Lamberov, S. R. Egorova, I. R. Ilyasov, K. K. Gilma- 134, 244509; b) R. A. van Santen, M. Neurock, S. G. Shetty,
nov, S. V. Trifonov, V. M. Shatilov, A. S. Ziyatdinov, Kinet. Chem. Rev. 2010, 110, 2005 – 2048.
Catal. 2007, 48, 136 – 142; b) N. A. Khan, A. Uhl, S. Shaikhut- [314] a) H.-Y. Su, Y. Gorlin, I. C. Man, F. Calle-Vallejo, J. K.
dinov, H. J. Freund, Surf. Sci. 2006, 600, 1849 – 1853. Nørskov, T. F. Jaramillo, J. Rossmeisl, Phys. Chem. Chem.
[299] a) D. Teschner, J. Borsodi, Z. Kis, L. Szentmikl si, Z. Rvay, A. Phys. 2012, 14, 14010 – 14022; b) R. R. Chianelli, G. Berhault, P.
Knop-Gericke, R. Schlçgl, M. Scheffler, D. Torres, P. Sautet, J. Raybaud, S. Kasztelan, J. Hafner, H. Toulhoat, Appl. Catal. A
Phys. Chem. C 2010, 114, 2293 – 2299; b) D. Teschner, Z. Revay, 2002, 227, 83 – 96; c) S. Wang, V. Petzold, V. Tripkovic, J. Kleis,
J. Borsodi, M. Hvecker, A. Knop-Gericke, R. Schlçgl, D. J. G. Howalt, E. Skulason, E. M. Fernandez, B. Hvolbaek, G.
Milroy, S. D. Jackson, D. Torres, P. Sautet, Angew. Chem. Int. Jones, A. Toftelund, H. Falsig, M. Bjorketun, F. Studt, F. Abild-
Ed. 2008, 47, 9274 – 9278; Angew. Chem. 2008, 120, 9414 – 9418. Pedersen, J. Rossmeisl, J. K. Nørskov, T. Bligaard, Phys. Chem.
[300] a) M. Armbrster, R. Schlçgl, Y. Grin, Sci. Technol. Adv. Mater. Chem. Phys. 2011, 13, 20760 – 20765.
2014, 15, 034803 – 034821; b) R. Leary, Z. Saghi, M. Armbrus- [315] a) J. K. Nørskov, F. Abild-Pedersen, F. Studt, T. Bligaard, Proc.
ter, G. Wowsnick, R. Schlçgl, J. M. Thomas, P. A. Midgley, J. Natl. Acad. Sci. USA 2011, 108, 937 – 943; b) C. J. Cramer, D. G.
Phys. Chem. C 2012, 116, 13343 – 13352; c) M. Armbrster, K. Truhlar, Phys. Chem. Chem. Phys. 2009, 11, 10757 – 10816; c) J.
Kovnir, M. Friedrich, D. Teschner, G. Wowsnick, M. Hahne, P. Greeley, M. Mavrikakis, Nat. Mater. 2004, 3, 810 – 815.
Gille, L. Szentmiklosi, M. Feuerbacher, M. Heggen, F. Girgs- [316] A. J. Medford, J. Wellendorff, A. Vojvodic, F. Studt, F. Abild-
dies, D. Rosenthal, R. Schlçgl, Y. Grin, Nat. Mater. 2012, 11, Pedersen, K. W. Jacobsen, T. Bligaard, J. K. Nørskov, Science
690 – 693; d) M. Armbrster, M. Behrens, F. Cinquini, K. 2014, 345, 197 – 200.

3518 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520
 Angewandte

15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Heterogeneous Catalysis Chemie

[317] a) J. J. Mortensen, L. B. Hansen, B. Hammer, J. K. Nørskov, J. [333] F. J. Melendez, A. Degollado, M. E. Castro, N. A. Caballero,
Catal. 1999, 182, 479 – 488; b) L. M. Aparicio, J. A. Dumesic, H. J. A. Guevara-Garcia, T. Scior, Inorg. Chim. Acta 2014, 420,
Topsøe, M. Boudart, J. K. Nørskov, Ammonia synthesis kinet- 149 – 158.
ics: surface chemistry, rate expressions, and kinetic analysis, J. C. [334] a) J. Wang, C. Y. Fan, K. Jacobi, G. Ertl, Surf. Sci. 2001, 481,
Baltzer AG, 1994. 113 – 118; b) Y. D. Kim, H. Over, G. Krabbes, G. Ertl, Top.
[318] a) F. Bozso, G. Ertl, M. Grunze, M. Weiss, J. Catal. 1977, 49, 18 – Catal. 2001, 14, 95 – 100; c) H. Over, Y. D. Kim, A. P. Seitsonen,
41; b) F. Boszo, G. Ertl, M. Weiss, J. Catal. 1977, 50, 519 – 529; S. Wendt, E. Lundgren, M. Schmid, P. Varga, A. Morgante, G.
c) S. R. Bare, D. R. Strongin, G. A. Somorjai, J. Phys. Chem. Ertl, Science 2000, 287, 1474 – 1476; d) D. W. Goodman,
1986, 90, 4726 – 4729. C. H. F. Peden, M. S. Chen, Surf. Sci. 2007, 601, L124 – L126;
[319] a) A. Vojvodic, J. K. Nørskov, F. Abild-Pedersen, Top. Catal. e) F. M. Hoffmann, M. D. Weisel, C. H. F. Peden, Surf. Sci.
2014, 57, 25 – 32; b) Q. Ge, M. Neurock, Chem. Phys. Lett. 2002, 1991, 253, 59 – 71.
358, 377 – 382; c) A. Ruban, B. Hammer, P. Stoltze, H. L. [335] a) K. Reuter, M. Scheffler, Phys. Rev. B 2003, 68, 045407; b) K.
Skriver, J. K. Nørskov, J. Mol. Catal. A 1997, 115, 421 – 429. Reuter, M. Scheffler, Phys. Rev. B 2002, 65, 165403.
[320] S. G. Wang, B. Temel, J. A. Shen, G. Jones, L. C. Grabow, F. [336] D. Rosenthal, F. Girgsdies, O. Timpe, G. Weinberg, R. Schlçgl,
Studt, T. Bligaard, F. Abild-Pedersen, C. H. Christensen, J. K. Z. Phys. Chem. 2011, 225, 57 – 68.
Nørskov, Catal. Lett. 2011, 141, 370 – 373. [337] R. Blume, M. Hvecker, S. Zafeiratos, D. Teschner, A. Knop-
[321] a) J. R. Kitchin, J. K. Nørskov, M. A. Barteau, J. G. Chen, Phys. Gericke, R. Schlçgl, L. Gregoratti, A. Barinov, M. Kiskinova in
Rev. Lett. 2004, 93, 156801; b) S. Zafeiratos, S. Piccinin, D. Nanostructured Catalysts: Selective Oxidation, Vol. 19 (Eds.: C.
Teschner, Catal. Sci. Technol. 2012, 2, 1787 – 1801; c) T. Con- Hess, R. Schlçgl), The Royal Society of Chemistry, Cambridge,
tant, A. M. Karim, V. Lebarbier, Y. Wang, F. Girgsdies, R. 2011, pp. 248 – 265.
Schlçgl, A. Datye, J. Catal. 2008, 257, 64 – 70; d) J. A. Rodri- [338] a) T. C. R. Rocha, M. Havecker, A. Knop-Gericke, R. Schlçgl,
guez, M. Kuhn, J. Phys. Chem. 1996, 100, 381 – 389; e) R. J. Catal. 2014, 312, 12 – 16; b) Y. Lei, F. Mehmood, S. Lee, J.
Schlçgl, G. Loose, M. Wesemann, A. Baiker, J. Catal. 1992, 137, Greeley, B. Lee, S. Seifert, R. E. Winans, J. W. Elam, R. J.
139 – 157; f) A. Baiker, D. Gasser, J. Lenzner, A. Reller, R. Meyer, P. C. Redfern, D. Teschner, R. Schlçgl, M. J. Pellin,
Schlçgl, J. Catal. 1990, 126, 555 – 571. L. A. Curtiss, S. Vajda, Science 2010, 328, 224 – 228; c) D. S. Su,
[322] a) P. Strasser, S. Koh, T. Anniyev, J. Greeley, K. More, C. F. Yu, T. Jakob, T. W. Hansen, D. Wang, R. Schlçgl, B. Freitag, S.
Z. C. Liu, S. Kaya, D. Nordlund, H. Ogasawara, M. F. Toney, A. Kujawa, Angew. Chem. Int. Ed. 2008, 47, 5005 – 5008; Angew.
Nilsson, Nat. Chem. 2010, 2, 454 – 460; b) A. Logadottir, J. K. Chem. 2008, 120, 5083 – 5086; d) V. I. Bukhtiyarov, A. I.
Nørskov, Surf. Sci. 2001, 489, 135 – 143; c) J. M. Schultz, J. Catal. Nizovskii, H. Bluhm, M. Hvecker, E. Kleimenov, A. Knop-
1972, 27, 64 – 69. Gericke, R. Schlçgl, J. Catal. 2006, 238, 260 – 269; e) V. V.
[323] A. J. Medford, A. C. Lausche, F. Abild-Pedersen, B. Temel, Kaichev, V. I. Bukhtiyarov, M. Hvecker, A. Knop-Gericke, R.
N. C. Schjodt, J. K. Norskov, F. Studt, Top. Catal. 2014, 57, 135 – Schlçgl, Kinet. Catal. 2003, 44, 432 – 440; f) V. I. Bukhtiyarov,
142. M. Hvecker, V. V. Kaichev, A. Knop-Gericke, R. W. Mayer, R.
[324] J. S. Hummelshøj, F. Abild-Pedersen, F. Studt, T. Bligaard, J. K. Schlçgl, Phys. Rev. B 2000, 67, 235422; g) Y. Uchida, X. Bao, K.
Nørskov, Angew. Chem. Int. Ed. 2012, 51, 272 – 274; Angew. Weiss, R. Schlçgl, G. Ertl, Surf. Sci. 1998, 401, 469 – 475; h) T.
Chem. 2012, 124, 278 – 280. Schedel-Niedrig, X. Bao, M. Muhler, R. Schlçgl, Ber. Bunsen-
[325] a) M. Armbrster, M. Behrens, F. Cinquini, K. Fçttinger, Y. Ges. 1997, 101, 994 – 1006; i) X. Bao, M. Muhler, B. Pettinger, Y.
Grin, A. Haghofer, B. Klçtzer, A. Knop-Gericke, H. Lorenz, A. Uchida, G. Lehmpfuhl, R. Schlçgl, G. Ertl, Catal. Lett. 1995, 32,
Ota, S. Penner, J. Prinz, C. Rameshan, Z. Revay, D. Rosenthal, 171 – 183; j) B. Pettinger, X. Bao, I. Wilcock, M. Muhler, R.
N. Rupprechter, P. Sautet, R. Schlçgl, L. Shao, L. Szentmiklosi, Schlçgl, G. Ertl, Angew. Chem. Int. Ed. Engl. 1994, 33, 85 – 86;
D. Teschner, D. Torres, R. Wagner, R. Widmer, G. Wowsnick, Angew. Chem. 1994, 106, 113 – 115; k) X. Bao, M. Muhler, B.
ChemCatChem 2012, 4, 1048 – 1063; b) D. Teschner, J. Borsodi, Pettinger, R. Schlçgl, G. Ertl, Catal. Lett. 1993, 22, 215 – 225;
Z. Kis, L. Szentmiklosi, Z. Revay, A. Knop-Gericke, R. Schlçgl, l) A. Nagy, G. Mestl, Appl. Catal. A 1999, 188, 337 – 353;
D. Torres, P. Sautet, J. Phys. Chem. C 2010, 114, 2293 – 2299. m) R. B. Grant, R. M. Lambert, Surf. Sci. 1984, 146, 256 – 268;
[326] V. Viswanathan, H. A. Hansen, J. Rossmeisl, J. K. Nørskov, n) C. Backx, C. P. M. Degroot, P. Biloen, W. M. H. Sachtler,
ACS Catal. 2012, 2, 1654 – 1660. Surf. Sci. 1983, 128, 81 – 103; o) R. W. Joyner, M. W. Roberts,
[327] a) P. Mani, R. Srivastava, P. Strasser, J. Phys. Chem. C 2008, 112, Chem. Phys. Lett. 1979, 60, 459 – 462.
2770 – 2778; b) S. Koh, P. Strasser, J. Am. Chem. Soc. 2007, 129, [339] T. E. Jones, T. C. R. Rocha, A. Knop-Gericke, C. Stampfl, R.
12624 – 12625. Schlçgl, S. Piccinin, Phys. Chem. Chem. Phys. 2014, 16, 9002 –
[328] I. C. Man, H. Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I. 9014.
Martinez, N. G. Inoglu, J. Kitchin, T. F. Jaramillo, J. K. Nørskov, [340] A. J. Nagy, G. Mestl, R. Schlçgl, J. Catal. 1999, 188, 58 – 68.
J. Rossmeisl, ChemCatChem 2011, 3, 1159 – 1165. [341] a) W. X. Li, C. Stampfl, M. Scheffler, Phys. Rev. B 2003, 68,
[329] a) A. A. Topalov, I. Katsounaros, M. Auinger, S. Cherevko, J. C. 045407; b) W. X. Li, C. Stampfl, M. Scheffler, Phys. Rev. Lett.
Meier, S. O. Klemm, K. J. J. Mayrhofer, Angew. Chem. Int. Ed. 2003, 90, 066104.
2012, 51, 12613 – 12615; Angew. Chem. 2012, 124, 12782 – [342] a) C. Rehren, M. Muhler, X. Bao, R. Schlçgl, G. Ertl, Z. Phys.
12785; b) K. Schlçgl, K. J. J. Mayrhofer, M. Hanzlik, M. Chem. 1991, 174, 11 – 52; b) L. Lefferts, J. G. Vanommen,
Arenz, J. Electroanal. Chem. 2011, 662, 355 – 360. J. R. H. Ross, Appl. Catal. 1987, 34, 329 – 339; c) L. Lefferts,
[330] X. G. Wang, W. Weiss, S. K. Shaikhutdinov, M. Ritter, M. J. G. Vanommen, J. R. H. Ross, Appl. Catal. 1987, 31, 291 – 308;
Petersen, F. Wagner, R. Schlçgl, M. Scheffler, Phys. Rev. Lett. d) L. Lefferts, J. G. Vanommen, J. R. H. Ross, Appl. Catal.
1998, 81, 1038 – 1041. 1986, 23, 385 – 402.
[331] K. H. Schwarz, P. Blaha in Fifth World Congress on Compu- [343] a) P. J. Van den Hoek, E. J. Baerends, R. A. Vansanten, J. Phys.
tional Mechanics (Eds.: H. A. Mang, F. G. Rammerstorfer, J. Chem. 1989, 93, 6469 – 6475; b) C. Backx, J. Moolhuysen, P.
Eberhardsteiner), Proceedings, Vienna, 2002. Geenen, R. A. Vansanten, J. Catal. 1981, 72, 364 – 368.
[332] M. Saeys, M. F. Reyniers, J. W. Thybaut, M. Neurock, G. B. [344] V. I. Bukhtiyarov, M. Hvecker, V. V. Kaichev, A. Knop-
Marin, J. Catal. 2005, 236, 129 – 138. Gericke, R. W. Mayer, R. Schlçgl, Phys. Rev. B 2003, 67,
235422.

Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 3519
 15213773, 2015, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.201410738 by Ruhr-Universität Bochum, Wiley Online Library on [26/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
.
Angewandte R. Schlçgl
Reviews

[345] S. Linic, J. Jankowiak, M. A. Barteau, J. Catal. 2004, 224, 489 – [352] a) A. Guram, A. Hagemeyer, C. G. Lugmair, H. W. Turner,
493. A. F. Volpe, Jr., W. H. Weinberg, K. Yaccato, Adv. Synth. Catal.
[346] a) N. Schweitzer, H. Xin, E. Nikolla, J. T. Miller, S. Linic, Top. 2004, 346, 215 – 230; b) L. Chen, E. G. Derouane, J. C. Vedrine,
Catal. 2010, 53, 348 – 356; b) P. Christopher, S. Linic, Chem- Appl. Catal. A 2004, 270, 157 – 163.
CatChem 2010, 2, 78 – 83. [353] a) A. Tompos, J. L. Margitfalvi, L. Vegvari, A. Hagemeyer, T.
[347] a) S. Zander, E. L. Kunkes, M. E. Schuster, J. Schumann, G. Volpe, C. J. Brooks, Top. Catal. 2010, 53, 100 – 107; b) M.
Weinberg, D. Teschner, N. Jacobsen, R. Schlçgl, M. Behrens, Woodhouse, B. A. Parkinson, Chem. Soc. Rev. 2009, 38, 197 –
Angew. Chem. Int. Ed. 2013, 52, 6536 – 6540; Angew. Chem. 210.
2013, 125, 6664 – 6669; b) J. T. Kozlowski, M. Behrens, R. [354] a) C. Baltes, S. Vukojevic, F. Schth, J. Catal. 2008, 258, 334 –
Schlçgl, R. J. Davis, ChemCatChem 2013, 5, 1989 – 1997; c) S. 344; b) Q. Huo, D. I. Margolese, U. Ciesla, P. Feng, T. E. Gier, P.
Kaluza, M. Behrens, N. Schiefenhçvel, B. Kniep, R. Fischer, R. Sieger, R. Leon, P. M. Petroff, F. Schth, G. D. Stucky, Nature
Schlçgl, ChemCatChem 2011, 3, 189 – 199. 1994, 368, 317 – 321.
[348] A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J. O. [355] W. D. Pyrz, D. A. Blom, N. R. Shiju, V. V. Guliants, T. Vogt,
Muller, R. Schlçgl, J. M. Carlsson, J. Mater. Chem. 2008, 18, D. J. Buttrey, Catal. Today 2009, 142, 320 – 328.
4893 – 4908. [356] a) B. Cormary, F. Dumestre, N. Liakakos, K. Soulantica, B.
[349] a) T. Reier, M. Oezaslan, P. Strasser, ACS Catal. 2012, 2, 1765 –
Chaudret, Dalton Trans. 2013, 42, 12546 – 12553; b) U. Ciesla,
1772; b) G. Jones, J. G. Jakobsen, S. S. Shim, J. Kleis, M. P.
D. Demuth, R. Leon, P. Petroff, G. Stucky, K. Unger, F. Schth,
Andersson, J. Rossmeisl, F. Abild-Pedersen, T. Bligaard, S.
J. Chem. Soc. Chem. Commun. 1994, 1387 – 1388.
Helveg, B. Hinnemann, J. R. Rostrup-Nielsen, I. Chorkendorff,
[357] J. M. Thomas, Angew. Chem. Int. Ed. Engl. 1988, 27, 1673 –
J. Sehested, J. K. Nørskov, J. Catal. 2008, 259, 147 – 160.
1691; Angew. Chem. 1988, 100, 1735 – 1753.
[350] R. Schlçgl, Angew. Chem. Int. Ed. 1998, 37, 2333 – 2336; Angew.
[358] R. Schlçgl, Carbon in Catalysis in Advances in Catalysis, Vol. 56
Chem. 1998, 110, 2467 – 2470.
(Eds.: B. C. Gates, F. C. Jentoft), Elsevier, Amsterdam, 2013,
[351] S. B. Abd Hamid, R. Schlçgl, H. J. Fecht, M. Werner in The
Nano-Micro Interface: Bridging the Micro and Nano Worlds, pp. 103 – 185.
Wiley-VCH, Weinheim, 2004, S. 139 – 150.

3520 www.angewandte.org  2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 3465 – 3520