%0 Thesis %A Mitchell, Garrett M %D 2020 %T Morphology, Properties and Reactivity of Nanostructures %U https://hammer.purdue.edu/articles/thesis/Morphology_Properties_and_Reactivity_of_Nanostructures/12264830 %R 10.25394/PGS.12264830.v1 %2 https://hammer.purdue.edu/ndownloader/files/22589489 %K Catalysis %K Nanoparticles %K batteries %K STEM %K TEM %K Chemical Engineering not elsewhere classified %K Materials Engineering not elsewhere classified %K Catalysis and Mechanisms of Reactions %X
Metal nanoparticles have long been of paramount importance in many areas such as: emission reduction in cars, hydrogen production via the water-gas shift reaction, and lithium-ion storage in batteries. For these purposes, the size and shape of the nanoparticles have been shown to play a crucial role in improving nanoparticle performance.

For characterization of nanostructures, the use of transmission electron microscopy (TEM) has been shown to be extremely useful. Via a TEM instrument, one can learn about nanoparticle properties such as: particle size, 3D morphology, chemical composition, fine structure, crystallography, even to atomic resolution. No other technique boasts such ability at such a high xyz resolution. This work includes TEM work for many different applications within catalysis and energy storage fields.

In catalytic applications, the <1 nm particle sizes often sought after generally lead to higher activity per unit mass of the catalyst, but also have the tendency to sinter due to concomitant increases in the surface free energy, leading to catalyst deactivation especially at elevated temperatures. To investigate the sintering, (Pt,Au)-iron oxide heterodimer nanoparticles were heated in the microscope with simultaneous imaging. For that purpose, the sample was irradiated with a 532 nm pulsed laser, with laser powers of 4-25 mW within a TEM microscope to investigate particle sintering as it happens. The Au and Pt phases were both found to wet over the Fe3O4 phase, a behavior opposite to the Strong Metal Support Interactions (SMSI - caused by oxide wetting the metal) which were expected from well-known literature reports. This new behavior demonstrates that not only nanoparticle size, but also the support particle size can affect catalytic properties. This is shown by the fact that the size of the support oxide in these heterodimer nanoparticles is only 3 times the diameter of the active metal nanoparticles, compared to a greater than 20 times size difference for a standard metal oxide supported nanoparticle system.
Nanoparticle metal catalysts can also undergo significant catalytic improvement via the addition of promoting metals. Kinetics were measured on a series of Pt/Co on carbon nanotube support catalysts, and addition of Co was seen to improve the turnover frequency by 10 times. Leaching of the bulk Co phases, while preserving PtCo alloy structures, reduced activity by more than 18 times demonstrating the need for a Pt/CoOxHy interface for catalytic promotion, and showing that PtCo alloying did not produce the promotion effect.

Although, for the PtCo catalysts for WGS, the formation of a Co-oxyhydroxide phase was proved to be vital, nanoparticle alloying is also well-known to improve dehydrogenation kinetics. This was shown for a series of PtM catalysts with core/shell structures, which were found to be highly selective for propane dehydrogenation as a result of the PtM intermetallic phase. XAS studies of these materials led to the discovery that formation of a continuous PtM alloy surface layer that is 2–3 atomic layers thick was sufficient to obtain identical catalytic properties between those of the core–shell and full alloy catalysts. TEM characterization was also performed to determine the core/shell nature of these catalysts.

Another interesting morphological "tuning knob" of nanoparticle catalysts is related to Reactive metal–support interactions (RMSI). RMSI can have electronic, geometric and compositional effects that can be used to tune catalytic active sites. Generally, non-oxide supports are disregarded as unable to undergo RMSI. However, we report an example of non-oxide-based RMSI between platinum and Nb2CTx MXenes--a recently developed, two-dimensional metal carbide, with a dopant labeled as T. The surface functional groups can be reduced, and a Pt–M surface alloy is formed. WGS reaction kinetics reveal that these RMSI supports stabilize the relevant nanoparticles and generate higher H2O activation ability and thus higher rates compared with a non-reducible support or a bulk niobium carbide. This RMSI between platinum and the niobium MXene support can be extended to other members of the MXene family and opens new avenues for the facile design and manipulation of functional bimetallic nanoparticle catalysts.

Other important catalytic nanostructures are Au/TS-1 (Titanosilicalite-1, a zeolite with the MFI structure) catalysts which can be used to make propylene oxide (PO), an important industrial intermediate, and are extremely interesting due to the potential for one-pot chemical reactions, which will save on capital costs. The kinetics of propylene epoxidation over these Au/TS-1 catalysts were measured in a continuous stirred tank reactor (CSTR) free from temperature and concentration gradients. Apparent reaction orders were measured at 473 K for H2 (0.7 order), O2 (0.2), and C3H6 (0.2) for a series of Au/TS-1 catalysts with varied Au (0.02–0.09 wt%) and Ti (Si/Ti: 75–143) contents. These measured orders were consistent with those reported previously. Co-feeding propylene oxide enabled measurement of the apparent reaction order in propylene oxide (−0.4 to −0.8 order), showing, for the first time, and it was found that relevant pressures of propylene oxide reversibly inhibit propylene epoxidation over Au/TS-1, while co-feeding carbon dioxide and water has no effect on the propylene epoxidation rate. Analysis of previously proposed two-site reaction mechanisms in light of these new reaction orders for O2 (0.4), H2 (1), and C3H6 (0.4), corrected to account for propylene oxide inhibition, provides further evidence that propylene epoxidation over Au/TS-1 occurs via a simultaneous mechanism requiring two distinct, but adjacent, types of sites, and not by a sequential mechanism that invokes migration of H2O2 formed on Au sites to PO forming Ti sites. H2 oxidation rates are not inhibited by propylene oxide, implying that the sites required for hydrogen oxidation are distinct from those required for propylene epoxidation. Those who intend on performing kinetics in the future are encouraged to perform a simple conversion-based tau-test, outlined in the relevant chapter of this thesis, to determine whether products inhibit reaction rates.





Yet another important field in which nanoparticle morphology research is essential is that of development of lithium-ion batteries. The current commercial graphite anode for lithium batteries is unfortunately prone to formation of lithium plating during use, from which well-documented safety issues arise. We demonstrated the use of an alternative anode, antimony, to have a measured specific capacity that is 1.6x higher than the theoretical capacity of graphite. Antimony, however, suffers from low cyclability due to large volumetric changes (~150%) upon the expansion caused by lithiation. To combat this problem, several different synthesis methods to produce nanoparticles of differing structures were tested and it was found that amine boranes produce a unique 3D nanochain structure with stable particle sizes of ~30 nm. These “3D nanochains” were found to have a stable charge capacity retention (98%) after 100 cycles due to their unique morphology which accommodates the lithiation expansion.

The role of sulfur nanostructures in lithium–sulfur batteries was also examined. Carbon–sulfur composites without crystalline sulfur demonstrate a high specific capacity of ≈1000 Ah kg−1 after 100 cycles with a gravimetric current of 557 A kg−1. This high rate capacity is found to depend on sulfur distribution which, in turn, is controlled by the synthesis pathway.

In conclusion, the morphology of nanostructures affects many different aspects of performance, rate, and stability. Further study into these details are expected to generate additional knowledge of a wide variety of interesting nanomaterials.
%I Purdue University Graduate School