Preparation and Characterization of Cu/ZnO
Catalysts for Methanol Synthesis
vorgelegt von
Diplom-Chemiker
Stefan Zander (geb. Kißner)
aus Berlin
Von der Fakultät II - Mathematik und Naturwissenschaften
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Naturwissenschaften
Dr. rer. nat.
genehmigte Dissertation
Promotionsausschuss:
Vorsitzender: Prof. Dr. Reinhard Schomäcker
Berichter/Gutachter: Prof. Dr. Robert Schlögl
Berichter/Gutachter: Prof. Dr. Thorsten Ressler
Berichter/Gutachter: Prof. Dr. Martin Muhler
Tag der wissenschaftlichen Aussprache: 07.08.2012
Berlin, 2013
D 83
Abstract
i
Abstract
In this work, systematic investigations of the preparation of Cu/ZnO-based methanol synthesis
catalysts are presented. The catalyst precursors were prepared by co-precipitation, followed by
aging, filtrating, washing and drying according to the proven, but not completely understood
industrial preparation method. Subsequent calcination and reduction led to the catalytically
active Cu/ZnO/X catalyst, X being a refractory oxide acting as structural promoter. The
investigations focus on the chemistry of the zincian malachite precursor that was identified as
the material yielding the best catalytic performance with the aim of identifying the role of the
different synthesis parameters on its formation mechanism and properties and of establishing
structure-performance-relationships that explain the role of the synthesis conditions and the
structural promoter phase X on the final catalytic activity.
Co-precipitation (Cu:Zn = 70:30) was performed in a pH- and temperature-controlled (338 K)
manner and enabled homogeneous distribution of the metal ions in the amorphous initial
precipitate which transformed into crystalline zincian malachite during aging. This aging step
was found to be critical with regard to the incorporation of Zn into zincian malachite and was
investigated by in-situ methods. Therefore, it had to be decoupled from the prior co-
precipitation step using co-precipitation with continuous spray-drying. As a function of aging
pH, two different aging mechanisms were found that explain the effect of the synthesis
conditions in the early stages of preparation on the structural properties of the precursor and
later the resulting catalyst. Low pH-values (5.0-6.5) trigger a direct co-condensation
mechanism, while at high pH values (7.0-8.0) a transient sodium zinc carbonate phase was
observed upon crystallization of the zincian malachite precursor phase. The Zn incorporation
into the zincian malachite precursor phase was higher at low pH values. Temperature was found
to accelerate both pathways at a given pH value. Based on these results, the setting of the
synthesis parameters in the applied catalyst preparation method can be rationalized. They have
been optimized to yield maximal Cu,Zn substitution in zincian malachite which in turn is a
precondition for final nanostructuring of the catalyst.
Also in conventional batch synthesis of Cu/ZnO catalysts the application of different pH-values
in the range of pH 6.0-9.0 during co-precipitation was observed to influence the precursor
chemistry. Application of pH values ≥ 6.5 led to higher phase fraction of zincian malachite at
the expense of the undesired Zn-rich by-phase aurichalcite. As a consequence, more Zn was
inserted into zincian malachite after aging, leading to smaller CuO domain size in the calcined
catalyst. For pH-values in the basic regime, formation of two clearly different substituted
Abstract
ii
zincian malachite phases was found indicating inhomogeneous Zn distribution in the precursor
material. The pH-dependent switch of the aging mechanism observed during the previously
described in-situ experiments is a likely explanation for the differences in Zn incorporation.
However, the highest Cu surface area, which is a prerequisite for an efficient catalyst, was
obtained for catalysts prepared at pH 8.5. Unfortunately, we were not able to track back this
observation directly to the synthesis pH in a simple synthesis parameter–structure–performance
relationship. The batch process is probably more complex as variation of the parameter pH may
induce numerous changes in the precursor material that can lead to different and partially
compensating effects for the resulting catalyst.
ZnO is known to act as a spacer for the single Cu particles in the Cu/ZnO catalyst and to enable
the widely studied Cu-ZnO synergy which beneficially affects the activity. MgO was
investigated to act as a substitute for ZnO following the substituted malachite preparation
approach. At the same Cu content
(80 mol%)
, the geometric influence turned out to be even
better compared to ZnO but the synergetic effect of Cu and ZnO during methanol synthesis
from CO
2
/CO/H
2
was lacking. By subsequent impregnation with ZnO both geometric and
synergetic effects were combined in a Cu/MgO/ZnO catalyst which exhibited a higher activity
than Cu/ZnO and Cu/MgO. Thus, the geometric and synergetic effects of the oxide components
have been separated during synthesis. Interestingly, if the feed gas was changed to CO/H
2
,
Cu/MgO was by far most active.
The effect of Ga
2
O
3
as a promoter in the Cu/ZnO/(Ga
2
O
3
) system was investigated by preparing
a sample series with increasing Ga concentration. Ga contents up to 3 mol% were incorporated
in the zincian malachite precursor despite the charge mismatch and changed the characteristics
of the sample dramatically. After calcination, some of the Ga was incorporated in the ZnO.
After reduction, the Cu surface area was increased by 100% and the methanol synthesis activity
by 60% compared to the binary Cu/ZnO reference system. Higher Ga contents led to
segregation and inhomogeneous microstructure of the resulting catalyst. The functionality of Ga
promotion was found to critically depend on the homogeneous distribution of Ga. The best
distribution was achieved by incorporation into the zincian malachite precursor phase and a
linear correlation of the (Zn,Ga) content in this phase with the catalytic activity of the final
catalyst was observed.
Zusammenfassung
iii
Zusammenfassung
Systematische Untersuchungen der Präparation von Cu/ZnO-basierten Methanolsynthese-
Katalysatoren sind Gegenstand dieser Arbeit. Die Katalysatorpräkursoren wurden gemäß dem
etablierten aber nur unzureichend verstandenen industriellen Syntheseweg hergestellt, der aus
den Schritten Co-Fällung, Altern, Filtern, Waschen und Trocknen besteht. Calcinierung und
Reduktion führen schließlich zum aktiven Cu/ZnO/X Katalysator, wobei X ein
temperaturbeständiges Oxid darstellt, welches als struktureller Promotor fungiert. Die
Untersuchungen richten sich auf die Chemie des Zink-Malachit-Präkursors, der letztendlich zu
einer hohen katalytischen Aktivität führt. Dabei sollen die Einflüsse verschiedener
Syntheseparameter auf die Bildung und Eigenschaften des Zink-Malachits und Mikrostruktur-
Aktivitäts-Korrelationen untersucht werden, um die katalytische Aktivität mittels
Syntheseparameter und Promotorphase X erklären zu können.
Durch pH- und temperaturkontrollierte (338 K) Co-Fällung (Cu:Zn 70:30) wird eine homogene
Verteilung der Metallionen im anfänglich amorphen Fällungsprodukt erreicht, welches durch
Altern zu Zink-Malachit kristallisiert. Die Alterung wird als entscheidender Schritt für die
Einlagerung von Zink-Ionen im Zink-Malachit angesehen und wurde mit Hilfe von in-situ
Methoden untersucht. Dafür war eine Entkopplung von der vorausgehenden Co-Fällung nötig,
was durch Co-Fällung und kontinuierliche Sprühtrocknung realisiert wurde. Zwei verschiedene
Alterungsmechanismen wurden, abhängig vom pH-Wert, beobachtet, welche den Einfluss der
Syntheseparameter in den frühen Präparationsschritten auf strukturelle Eigenschaften des
Präkursors und letztlich des Katalysators erklären können. Kleine pH-Werte (5.0-6.5) führen zu
direkter Co-Kondensation, während für höhere pH-Werte (7.0-8.0) eine vorübergehende
Natrium-Zink-Carbonat-Phase beobachtet wurde, bevor die Kristallisation der Zink-Malachit-
Phase einsetzte. Bei kleinen pH-Werten konnte eine erhöhte Substitution von Cu-Ionen durch
Zn-Ionen im Zink-Malachit erreicht werden. Erhöhung der Temperatur führte bei gegebenem
pH-Wert zu einer Beschleunigung beider Mechanismen. Basierend auf diesen Erkenntnissen
kann die Einstellung der Parameter bei dem vorliegenden Präparationsprozess vorgenommen
werden. Ziel dabei ist eine möglichst große Einlagerung von Zink im Zink-Malachit, was
wiederum eine Vorbedingung für die spätere Nano-Strukturierung ist.
Der pH-Wert spielt auch während der Co-Fällung eine entscheidende Rolle für die
Eigenschaften des Cu,Zn-Präkursors (70:30) und wurde in zwei Serien von Batch-Synthesen im
Bereich von 6.0-9.0 variiert. Wurden pH-Werte ≥ 6.5 verwendet, konnte nach dem Altern ein
hoher Phasenanteil von Zink-Malachit neben der unerwünschten Zink-reichen Nebenphase
Loading more pages...