Size-selected nanoparticles (atomic clusters) represent novel reference materials for nano-environmental and biomedical studies as well as model systems for catalysis research. It is now over 30 years since “magic numbers” were discovered in the mass spectra of atomic clusters, and ever since that time it has been commonplace to see images of the atomic structures of these typically sub-5nm model nanostructures. However these images are generally archetypal structures or the results of computer modelling, i.e., very rarely have they emerged from direct experimental measurements. Nowadays the availability of aberration-corrected HAADF STEM is transforming our approach to this structure challenge [1,2]. I will address the atomic structures of size-selected Au clusters, deposited onto standard carbon TEM supports from a mass-selected cluster beam source. Specific examples considered are the “magic number clusters” Au20, Au55, Au309, Au561, and Au923. The results expose (i) the size range in which the bulk fcc motif emerges, (ii) the metastability of frequently observed structures, (iii) the nature of equilibrium amongst competing isomers, (iv) the cluster surface and core melting points as a function of size and (v) the temperature the clusters reach under e-beam irradiation; (iv) and (v) come from in situ heating stage measurements. The cluster beam approach is applicable to more complex nanoparticles too, such as oxides and sulphides, while the STEM method can also be applied to chemically synthesized colloidal nanoparticles [3,4].

A second major challenge in cluster science is to translate the beautiful physics and chemistry of clusters into applications, notably catalysis [5] and reference standards [6]. Compared with the (powerful) colloidal route, the nanocluster beam approach involves no ligands, particles can be size selected by the mass filter, and metal particles containing challenging combinations of metals can readily be produced. However, until now the cluster approach has been held back by the extremely low rates of metal particle production, of the order of 1 microgram per hour. This is sufficient for surface science studies but several orders of magnitude below what is desirable even for research-level realistic reaction studies. In an effort to address this scale-up challenge, I will discuss the development of a new kind of cluster beam source, the “Matrix Assembly Cluster Source” (MACS). Results obtained [5,7,8] suggest that cluster beam yields of grams per hour may ultimately be feasible; mg scale has been demonstrated. Some practical catalysis results will be presented. An abundant flux of clusters demands new methods of cluster storage; I will discuss a multilayer deposition approach which releases small platelets of supported clusters [9]. 

[1] Z.Y. Li et al, Nature 451 46 (2008). [2] Z.W. Wang and R.E. Palmer, Phys. Rev. Lett. 108 245502 (2012). [3] J. Liu et al, Ultramicroscopy (online 22 Nov 2016). [4] N. Jian and R.E. Palmer, J. Phys. Chem. C 119 11114 (2015). [5] P.R. Ellis et al, Faraday Discussions 188 39 (2016). [6] R.C. Merrifield, Z.W. Wang, R.E. Palmer and J.R. Lead, Environ. Sci. Technol. 47 12426 (2013). [7] ] R.E. Palmer, L. Cao and F. Yin, Rev. Sci. Instrum. 87 046103 (2016). [8] V.T.A. Oiko et al, J. Chem. Phys. 145 166101 (2016). [9] N. Jian, K. Bauer and R.E. Palmer, submitted.