Abstract: A plasma-enhanced atomic layer deposition (ALD) process using the Ag(fod)(PEt3) precursor [(triethylphosphine)(6,6,7,7,8,8,8-heptafluoro-2,2-dimethy1-3,5-octanedionate)silver(I)] in combination with NH3-plasma is reported. The steady growth rate of the reported process (0.24 +/- 0.03 nm/cycle) was found to be 6 times larger than that of the previously reported Ag ALD process based on the same precursor in combination with H-2-plasma (0.04 +/- 0.02 nm/cycle). The ALD characteristics of the H-2-plasma and NH3-plasma processes were verified. The deposited Ag films were polycrystalline face-centered cubic Ag for both processes. The film morphology was investigated by ex situ scanning electron microscopy and grazing-incidence small-angle X-ray scattering, and it was found that films grown with the NH3-plasma process exhibit a much higher particle areal density and smaller particle sizes on oxide substrates compared to those deposited using the H-2-plasma process. This control over morphology of the deposited Ag is important for applications in catalysis and plasmonics. While films grown with the H-2-plasma process had oxygen impurities (similar to 9 atom %) in the bulk, the main impurity for the NH3-plasma process was nitrogen (similar to 7 atom %). In situ Fourier transform infrared spectroscopy experiments suggest that these nitrogen impurities are derived from NH surface groups generated during the NH3-plasma, which interact with the precursor molecules during the precursor pulse. We propose that the reaction of these surface groups with the precursor leads to additional deposition of Ag atoms during the precursor pulse compared to the H-2-plasma process, which explains the enhanced growth rate of the NH3-plasma process.

Abstract: In this study, a new bottom-up approach is proposed to predict the crystal structure of the substrate-induced polymorph (SIP) of an archetypal molecular semiconductor. In spite of intense efforts, the formation mechanism of SIPs is still not fully understood, and predicting their crystal structure is a very delicate task. Here, we selected lead phthalocyanine (PbPc) as a prototypical molecular material because it is a highly symmetrical yet nonplanar molecule and we demonstrate that the growth and crystal structure of the PbPc SIPs can be templated by the corresponding physisorbed self-assembled molecular networks (SAMNs). Starting from SAMNs of PbPc formed at the solution/graphite interface, the structural and energetic aspects of the assembly were studied by a combination of in situ scanning tunneling microscopy and multiscale computational chemistry approach. Then, the growth of a PbPc SIP on top of the physisorbed monolayer was modeled without prior experimental knowledge, from which the crystal structure of the SIP was predicted. The theoretical prediction of the SIP was verified by determining the crystal structure of PbPc thin films using X-ray diffraction techniques, revealing the formation of a new polymorph of PbPc on the graphite substrate. This study clearly illustrates the correlation between the SAMNs and SIPs, which are traditionally considered as two separate but conceptually connected research areas. This approach is applicable to molecular materials in general to predict the crystal structure of their SIPs.