Research at ITQ
1. Molecular Dynamics of the Diffusion of Hydrocarbons in Zeolites.
We use an atomistic approach based on forcefields to simulate how hydrocarbons diffuse through the microporous structure of zeolites. I employ the general purpose DLPOLY code which I recommend for it is robust, portable, easily modifyable, and quick in serial and parallel machines.
The MCM-22 structure showing the general
features of molecules diffusing through it. (A) A molecule diffuses through the
10 MR sinusoidal system. (B) Potential energy minima in the sinusoidal system.
(C) Potential minima in the large cavities. (D) Intracage motion in the large
cavities around a minimum. (E) Intracage motion around the two minima in the
large cavity. (F) Intercage diffusion through the large cavities.
The simulations allow us to calculate diffusion properties, very especially
the magnitude called "diffusivity", and also to visualise trajectories of
the individual molecules through the intricated microporous zeolite. We can
see whether the molecule/s remain in a restricted part of the solid, how
do they diffuse in the different channels and cavities, the time spent in
each channel, the individual diffusion coefficients through each channel,
the interactions between molecules in each channel or in the global structure,
the activation energies or energy profiles for each molecule, and the radial
distribution funtions among other magnitudes. So we can get a clear picture
of the physics of the process. Then, we establish comparisons of a given
sorbate in different zeolites, or different sorbates in one zeolite, or we
can study the effect of loading, or the effect of temperature, or we can
compare diffusion in single component and mixed component systems.
Diffusion of 2-methyl-hexane in MCM-22.
Molecules 1-6 are located in the
10 MR channels, and molecules 7-12 are
located in the 12 MR supercage systems.
Diffusion of n-heptane in MCM-22.
Molecules 1-6 are located in the
10 MR channels, and molecules 7-12 are
located in the 12 MR supercage systems.
2. Quantum Chemistry Studies of Titanio- and Germanio-zeolites.
We study the new properties acquired by a silica framework
when Ti and Ge atoms are introduced into it. Some of the new properties can be
explained by the great distortion caused in the framework after the Ti/Ge replaces
the original Si atom. This depends on how the structure can accommodate the
strain, so different structures are expected to yield different properties
-such as Lewis acidity- depending on its flexibility.
We use standard HF and DFT methods under Gaussian95 by using basis sets of
different qualities such as 3-21G**, 6-31G, 6-311-G, 6-31G**, among others,
and clusters of medium size.
Also we intend to explore the reactivity of these clusters in reactions
involving charge transfer.
Regarding Ge-zeolites, their interest comes not only from the heteroatom
substitution and the new properties brought in, but also from the zeolite
synthesis viewpoint. Introduction of Ge in the synthesis gel has led to
the discovery of new structural types such as ASV, BEC, IWR, IWW and UTL
among others, most of them containing double four rings as secondary building
unit (one notable exception being PKU-9). This is important, also from the point
of view that Ge stabilises low TOT angles in the structure, with the recognition
that certain topologies are more prone than others to bear such low angles.
A point raised by many researchers is whether low TOT angles are associated
with preferential sizes of rings and this is a point we have addressed in recent
research. A number of studies on this topic are underway.
Cluster used in the calculations with Ti-zeolites
3. Study of electric fields in microporous materials.
Electric charges are known to be responsible of nucleophilic and electrophilic
attack and in general, chemical reactivity is explained in terms of electric
charges. Organic molecules, when they react catalytically inside a microporous
material are immersed in an electric field that can modify its chemical behaviour.
In spite of this widely acknowledged principle, little has been reported on this
topic and that is why we have conducted (in collaboration with Dr. Dewi Lewis)
calculation on electric fields created by zeolites. We have been able to
demonstrate how electric fields created by the microporous framework does
influence the Broensted acidity and have found a correlation with the square
of the OH stretching frequency as shown in the figure below.
Correlation between electric field and OH stretching frequency
The circles in the figure above correspond to structures of AFI and CHA with
different chemical composition (pure silica and AlPO compositions), and also the
four crystallographic different protons are represented, so we see clearly how
independently of the source of the proton, there is a correlation between its
stretching frequency and the electric field, which is related to the energy
necessary to release the proton from its equilibrium geometry to a larger
OH distance in the direction of the OH bond.
4. Broensted acidity of zeolites and SAPOs.
A long standing research project has been carried out by our group on this
subject by using several computational techniques. A remarkable finding has
been made on several zeolites and related microporous materials which seems
to be of general application and it is the capability of atomistic forcefield
methodologies to simulate OH stretching frequencies, that characterise Broensted
acidity. One of the examples considered has been the zeolite ITQ-7 shown below.
View of ITQ-7 zeolite (whose IZA code is ISV)
In a purely siliceous structure an Aluminium atom has been introduced in each of the
crystallographic T-sites, and for each of them the four bridging protons have been
considered. A model like that was not able to reproduce the experimental infrared
spectrum for ITQ-7. When the template was considered into the structure, the Aluminium
distributions changed dramatically their relative stability. Considering first the
Al incorporation (with the template) and then -second- the proton incorporation,
the populations of the acid centres were recalculated and from that the infrared
calculated bands were recalculated. The resultant comparison with experiments shows
a much better agreement and it can be seen that both infrared bands are reproduced by our
calculations as well as their relative populations.
Experimental infrared bands of H-ITQ-7. The deconvolution
of the experimental is also shown. The calculations give two bands of 3577
and 3624 wavenumbers, in close agreement with the experimental at 3595 and 3624.
5. Topological and geometrical characterisation of solids.
Zeolites and related microporous solids are three-dimensional four-connected nets
where knots are tetrahedral atoms (Si,Al,P,Ge,...). A surprise comes when by only
considering a few combination rules, more than 140 different materials have been
characterised to this date, and hypotethical structures (that could be synthesised)
are virtually countless. A part of the secret of this amazing variety comes from
the flexibility of T-O-T angles, and so different topologies can -by deforming
from equilibrium of dense structural binding shown by quartz and others- can
be accommodated with reasonable strain that does not break the structure. In fact
all the microporous structures are quite stable. Characterisation of new structures
comes from two points: topology and bond angles and distances. An automated
way of calculating all these many data from the unit cell was lacking and a fortran
code for this purpose has been developed by our group (in collaboration with
Julian Gale). 'zeoTsites', as our code is called calculates all the necessary
data and shows the results in an orderly and elegant format so that the special
features of each structure are catched at a glimpse. More on this can be seen in this
web page (zeoTsites: a software tool for zeolites).
6. Computer simulation of zeolite synthesis.
An interest does exist in synthesising new zeotype structures on the ground of
-among others- potentially important industrial applications. Whether this topic
has been traditionally regarded as a series of cooking recipes, the recent panorama
is quite much changed and a general understanding is growing steadily. Even more
recent is the rational use of computational tools in this respect and our approach
comes from the atomistic forcefield approach that we have been using successfully
to so many other research topics related to zeolites. The role of templates and
structure directing agents is being developed and the energy interaction with the
zeolite framework is being revised and their different contributions studied
separately to see, for example, how van der Waals and electrostatic factors
influence the final synthesised material. Our first study on this topic has
been the templating role of the trispirrolidinium cation in the synthesis of
Al-ZSM-18 zeolite. Our results have validated the technique used and also the
template location has been optimised: two positions for the cation were found and
the minimum in energy was shown to be the experimentally observed by XRD. Secondly,
the important result of predicting the Al distribution in this material was
found and its preferential location in the 3-T rings was an important result
from the calculations, which was justified in terms of the electrostatic interactions
of the cation with the negative charged brought in the zeolite framework by the
incorporation of Al atoms.
ZSM-18 with the trispyrrolidinium cation in its minimum energy
conformation as found by the calculations and in agreement with the XRD results.
7. Metal-organic frameworks for adsorption and catalysis.
New microporous materials, metal-organic frameworks, have made their appearance in the
chemistry scene very recently. Since their synthesis in Yaghi's group in 1999,
the interest for these materials has experienced an exponential growth.
A key aspect here is that bonds are covalent, hence strong and directional, and
topologies are simple. Further, the reticular synthesis approach developed means
that these materials can be designed on an aprioristic basis. In fact nowadays their
number approaches one million.
That said, it may seem difficult to study MOF materials from a general viewpoint,
and this is precisely one of the aims in our group. General principles may help to
rationalise their properties and we start by adsorption of molecular hydrogen. The
idea behind is the very old that hydrogen bonds unoccupied d-orbitals of metals.
It is all not that simple, and several contributions have to be considered. Our
approach is based on electronic-structure methods. We also wonder not only by the type
of metal playing in the interaction but also by the shielding effect exerted by part
of the organic ligand groups. The overal effect tends to be a weak physisorption and
this deserves special attention from the point of view of the quantum-chemistry approach
to be employed. Finally, considerations about the topology of the inorganic building units
will be used in order to clarify ranges of adsorption energy that can be expected in these
materials, together with considerations relating free volume with accessible surface area
and density.
