COMMUNICATION
www.rsc.org/chemcomm | ChemComm
1
3
In situ C DEPT-MRI as a tool to spatially resolve chemical
conversion and selectivity of a heterogeneous catalytic reaction
occurring in a fixed-bed reactor{
Belinda S. Akpa, Michael D. Mantle, Andrew J. Sederman and Lynn F. Gladden*
Received (in Cambridge, UK) 3rd February 2005, Accepted 16th March 2005
First published as an Advance Article on the web 11th April 2005
DOI: 10.1039/b501698c
The distortionless enhancement by polarisation transfer
DEPT) nuclear magnetic resonance (NMR) technique, com-
Therefore it is important to understand how this heterogeneity in
flow pattern within the reactor influences chemical conversion and
selectivity within the reactor. While illustrated here with respect to
a small-scale fixed-bed reactor, the issue of heterogeneity in
hydrodynamics and the impact of this on catalyst effectiveness and
selectivity is generic to most, if not all, reactor designs from large
scale reactors down to microchannel reactors; all of these being
topics of current interest within our group.
(
bined with magnetic resonance imaging (MRI), has been used
to provide the first in situ spatially-resolved and quantitative
measurement of chemical conversion and selectivity within a
13
fixed-bed reactor using natural abundance C NMR.
Chemical mapping within optically opaque three-dimensional (3D)
systems is of generic importance in the chemical and biological
sciences. Further, it is often the case that it is a species in relatively
low concentration that is of interest, thereby requiring an imaging
capability that can identify particular molecular species within a
large reservoir of other species. Our interest is in understanding the
coupling of hydrodynamics and reaction kinetics in porous media
and, in particular, fixed-bed catalytic reactors. Here, we demon-
strate how conventional MRI combined with a magnetic
resonance polarisation enhancement technique can be used to
spatially resolve both the conversion and selectivity characterising
a heterogeneous catalytic reaction occurring within a fixed-bed
reactor. This is achieved without the need to isotopically enrich
In studying a hydrocarbon reaction using NMR spectroscopy,
1
two approaches are typically used, namely H or C observation.
13
1
Whenever possible, observation of the H nucleus is the method of
choice thereby exploiting the 99.9% natural abundance and
1
inherent NMR sensitivity of the H nucleus. Whenever spatially
resolving a signal, the success of the experiment depends on the
signal-to-noise available, as this will determine the spatial
resolution that can be obtained. However, the disadvantage of
1
1
using H observation is that the H nucleus has a relatively narrow
1
chemical shift range which, combined with the large number of H
resonances present in a typical spectrum of a hydrocarbon reaction
and the broadening of these resonances as a result of decreased
spin–spin relaxation times arising from the high liquid–solid
interfacial area within the sample, gives rise to a large number of
overlapping resonances making spectral assignment and the
quantification of peak areas difficult. The alternative approach is
1
3
with C, and thus is a cost-effective method that may be used in
routine applications. The technique is illustrated with reference to
the competitive etherification and hydration reactions of 2-methyl-
+
2-butene (2M2B) (ESI{) occurring within a fixed bed of H ion-
1
3
13
to observe the C nucleus. By moving to C observation, spectral
exchange resin; this is a reaction of industrial significance with
13
assignment is easier because the C nucleus has a much wider
1
chemical shift range than H, thereby reducing the number of
respect to the production of oxygenates for use as gasoline
1
additives. The products of the etherification and hydration
overlapping peaks. Consequently, individual spectral peaks should
be more readily resolved. However, given that the natural
reactions are tert-amyl methyl ether (TAME) and tert-amyl
alcohol (TAOH) respectively.
13
abundance of the C nucleus is only 1.07% and its NMR
1
sensitivity is lower than that of H by a factor of 5870, there is
Magnetic resonance techniques have already met with consider-
able success in studying catalyst systems at size-scales up to that of
individual catalyst particles (~2–5 mm). NMR spectroscopy is a
well-established tool for the in situ study of catalytic processes for
the purposes of understanding the nature of active sites and surface
species. More recently, magnetic resonance techniques have
provided insight into mass transport processes and carbon
1
3
considerable loss of signal-to-noise when employing C as
1
opposed to H observation. In solid state NMR, which typically
3
uses small, closed sample volumes (y1 cm ), this decrease in
sensitivity and natural abundance is addressed by isotopically
1
3
enriching the species of interest with C. However, in studying
chemical composition within a reactor operating under conditions
of continuous flow, this approach is prohibitively expensive.
To date, there have only been two reports of spatially mapping
conversion within a catalytic reactor. Whilst successful in
developing this field of research, both illustrate the limitations of
2,3
laydown within catalyst particles. However, in industrial-scale
applications the catalyst particles will, typically, be packed within
fixed-bed reactors. Magnetic resonance flow imaging has revealed
significant heterogeneity in the flow field within such reactors. For
example, in some regions of the inter-particle space, fluid flow may
1
4
using H observation as a generic approach. In the first report, we
be an order of magnitude faster than in other regions of the bed.
used volume selective magnetic resonance spectroscopy to quantify
the conversion of an esterification reaction occurring within a fixed
5
bed of ion-exchange resin. Conversion was quantified by the
1
chemical shift of the H resonance of the OH groups present in the
{
Electronic supplementary information (ESI) available: competitive
reactions of 2M2B. See http://www.rsc.org/suppdata/cc/b5/b501698c/
*
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 2741–2743 | 2741