Symposium SF01—Advanced Atomic Layer Deposition and Chemical Vapor Deposition Techniques and Applications

1D Nanosynthesis
Despite the massive progress achieved in the growth of carbon nanotube (CNT) forests on substrate, apart from lithographic patterning of the catalyst, little has been done to selectively (locally) control CNT height. Varying process parameters, gases, catalysts, or underlayer materials uniformly affects CNT height over the whole substrate surface. We show here how we can locally control CNT height, from no CNTs to up to 4X the nominal CNT height from iron catalyst on alumina underlayer by patterning reservoirs or by using overlayers during annealing or growth.
We show how we can lithographically pattern the reservoir material and obtain four different CNT heights during a unique growth process: no growth (Cu/Ag reservoir), nominal growth (no reservoir), 2X growth (Fe reservoir), and 4X growth (Mo reservoir).
Using a complementary technique, we show how a copper or nickel overlayer (stencil or bridge) placed above the catalyst surface during pre-annealing or during CNT growth deactivates the catalyst. We showed how we could pattern regions with CNTs and without CNTs by simply annealing the sample with a patterned overlayer positioned above its surface. We thus synthesized patterned CNT forests using a simple process, without the need for lithography.
We can combine the overlayer technique with one of the above-mentioned reservoirs (no reservoir, Cu/Ag reservoir, Fe reservoir, or Mo reservoir) to further modulate CNT growth by offsetting some or all of the growth enhancements achieved using the reservoirs. This modulation of the CNT height is a significant improvement compared to the "CNTs (one height) / no CNTs" patterning that has been achieved using lithography of the catalyst, and moves us closer to building 3D architectures of CNTs that could be useful for future electronic devices or sensors.

2D Nanosynthesis
Since the excitement about graphene, a monolayer of graphite, with its 2010 Nobel Prize, there has been extensive research in the synthesis of other non-carbon few/mono-layers exhibiting a variety of bandgaps and semiconducting properties (e.g., n or p type). The main approaches to deposit few/monolayers on a substrate are: (a) bottom-up synthesis from precursors using chemical vapor deposition (CVD) or (b) top-down exfoliation (liquid or mechanical) of bulk layered material. Using a Lego approach of superposing monolayers, we can envisage the fabrication of heterojunctions with original electronic behavior.
Here we show a combined bottom-up and top-down approach where (a) we synthesize in one step high yields of bulk layered materials by annealing a metal in the presence of a gas precursor (sulfur, phosphorous, or selenium) using chemical vapor deposition (CVD) and (b) we exfoliate and deposited few/mono-layers on a substrate from a sonicated mixture of our material in a specific solvent. It is interesting to note that, besides the structure being 2D layered, the properties of the nanomaterials synthesized slightly differ from the materials with the same stoichiometry synthesized using conventional chemical methods (e.g., solvothermal).
In this talk, we will discuss the chemical synthesis, the very extensive characterizations, and the lessons we learned in making multiple metal sulfides (Cu-S, Ag-S, Ni-S), metal phosphides (Ni-P, Cu-P), and metal selenides (Ag-Se, Cu-Se, W-Se, Mo-Se). We will see how we integrated these new materials into electrochemical devices and sensors.

Chemical vapor deposition (CVD) has been used extensively for synthesizing two-dimensional (2D) materials. Among various CVD methods, atmosphere pressure CVD is of great interest due to its low cost and promise for high-quality monolayer 2D materials. Despite extensive efforts, the understanding of the reaction mechanism and the key controlling parameters is still in its embryonic stage. Hence, it is extremely challenging to achieve large-scale continuous films. Here, we present a fluid-guided growth strategy and use MoSe₂ as a model system for understanding and controlling the growth of 2D materials. We have integrated experiment and computational fluid dynamics (CFD) analysis in the full-reactor scale. We have identified three key parameters in obtaining large-scale continuous film: good precursor mixing, low fluid velocity, and low shear rate near the substrate. Following these principles, we modified the geometry of the growth setup and successfully obtained inch-scale monolayer MoSe₂. This unprecedented integration of fluid design and simulation lays the foundation for designing new CVD systems to synthesize 2D materials and other thin-film nanomaterials.

Flexible Hybrid Electronics (FHE) and the Fan-Out redistribution layer (RDL) rely on electroplated Cu. In this work, we demonstrate direct low-temperature (~ 100 °C) plasma enhanced chemical vapor deposition (PECVD) of graphene on electroplated Cu over polyimide substrate, which passivates and strengthens the electroplated Cu circuit on polyimide. We investigate the effect of the H₂/CH₄ ratio on PECVD graphene growth and find that not only the quality of graphene but also the durability of Cu are affected. Folding for 100,000 cycles with a bending radius of 2.5 mm and the corresponding resistance tests show that the Cu circuit when covered with graphene grown with higher H₂/CH₄ ratios could sustain many more bending cycles. In addition, we find that graphene coverage could suppress the formation of Cu oxide under ambient environment for at least 8 weeks after the PECVD process.

Chemical vapor deposition (CVD) is the most widely used process to synthesize graphene for device applications. In this process, the interaction between graphene and the metal catalyst plays a critical role in the properties and quality of graphene, producing strains, wrinkles, moiré patterns, and in some cases an electronic bandgap in graphene. A very robust and repeatable phenomenon observed after graphene CVD is the modification of the substrate’s surface topography. Orientation-dependent large, faceted morphologies have been observed under graphene on all metal substrates. These facets grow by step-bunching and have a typical size of 50-100 nm. Previous studies are limited to a qualitative description of step bunching, and model the phenomenon as a bending energy-driven process. In this presentation, we present an approach to quantitatively characterize and analyze the facets’ orientation. We use electron backscatter diffraction (EBSD) in conjunction with atomic force microscopy to calculate the crystallographic orientation of facets that have step-like and pyramid-like topologies. We will present new approaches to visualize the crystallographic evolution of these steps using inverse pole figures (IPF). We carry statistical analysis on the observed step sizes and orientations on various catalyst grain orientations. We correlate the geometrical facet characterization method to measurements made by Transmission Electron Microscopy. Our studies reveal that the step-like facets comprise a high-symmetry low-index surface on one side and a vicinal surface on the other side, showing that system energy minimization is not the only driving force of the phenomena. We believe that faceting occurs as a result of a trade-off between the diffusion kinetics and the interfacial energy of the complete system which includes van der Waal’s interaction, the thermal strains in addition to the bending energy of graphene. MD simulations also show suppression of the substrate’s melting under graphene, which we believe plays a role in the kinetics of the faceting process. Our studies provide insights into metal-graphene interfaces that will enable us to better control graphene during synthesis.

Metal-oxides such as ZnO or Al₂O₃ synthesized through Atomic Layer Deposition (ALD) have been of great research interests as the candidate materials for ultra-thin tunnel barrier layers is. In this study, we have applied a 3D on-lattice Kinetic Monte Carlo (kMC) code developed by Timo Weckman’s group to simulate the growth mechanisms of the tunnel barrier layer and to evaluate the role of various experimentally relevant factors of the ALD processes. We have systematically studied the effect of a wide range of ALD parameters such as the chamber pressure and temperature, pulse/purge times, as well as the coverage of wetting layers on the substrate. The database generated from the kMC simulations was subsequently used as descriptors in the subsequent analyses via Machine Learning (ML) algorithms to gain more insights into the dynamics of the deposition processes. The results of a combined approach of kMC and ML were then compared to the experimental results. The support from NSF (EPMD Division) Award No. 1809284 is gratefully acknowledged. We also thanks to NERSC for the computational support.

In this study, we have utilized the reactive molecular dynamics (MD) simulations to model the Atomic Layer Deposition (ALD) process that forms an ultra-thin film of a tunnel barrier made of amorphous alumina. We chose the reactive MD approach over the ab-initio molecular dynamics simulation used in previous studies due to its lower computational cost, its ability to model over a relatively longer simulation period and its capability to assess atomistic-based dynamics for a larger substrate. To further refine and parameterize the existing ReaxFF potentials, we used the Density Functional based Tight Binding (DFTB) -based in conjunction with the Python toolkit as implemented in the Amsterdam Modeling Suite (AMS) program package. The additional training sets were made from evaluations of the potential energy surface (PES) scans of various ALD-relevant species including H₂O, OH, and the ALD precursors trimethylaluminum (TMA) and Bis(20ethyl-1,3-cyclopentadien-1-yl) magnesium (C₁₄H₁₈Mg), as well as the bond-dissociation energies during the reactions that occur when they interact. We systematically evaluated the role of the various ALD precursors including the TMA and C₁₄H₁₈Mg and the water pulse toward the chemical reactions that take place on the surface during ALD. We additionally evaluated the role of experimentally- observable parameters including the operating temperature and precursor concentrations for the internal structure of the amorphous alumina/magnesium as the final deposition products. Lastly, we assessed the role of wetting layers as the means to improve the quality and performance of the tunnel barrier. The support of NSF (Grant No. 1809284) from the Electronics, Photonics and Magnetic Devices (EPMD) Program is gratefully acknowledged. We also thanks to NERSC for the computational support.

The thermal decomposition process of metalorganic precursors, which are frequently used in various thin film deposition techniques, can have a governing effect on the overall growth and reaction kinetics of the thin film deposition methods employed. Particularly in cases of more complex thin-film synthesis schemes, such as combinatorial approaches employing molecular beam epitaxy and chemical vapor deposition, or for the deposition of films with multi-elemental composition, such as ternary and quaternary complex oxide thin films with perovskite structure, a quantitative understanding of reaction pathways, activation energies, and by-product formation at the atomic scale are becoming key towards an improved understanding of the reaction kinetics, as well as providing insights towards tailoring ligand chemistry of the metalorganic precursor employed.
A recently proposed thin film growth technique, dubbed hybrid molecular beam epitaxy,¹ has shown dramatic improvement in the film quality of complex perovskite oxide films in general and SrTiO₃ in particular, attributed to the accessibility of a self-regulated growth mechanism enabled by the volatility and decomposition characteristics of metalorganic precursors involved in the growth. Despite remarkable advancements in material quality, no clear understanding of the underlying atomic-scale growth mechanism has been achieved. For the particular case of SrTiO₃, where Sr is supplied by a thermal effusion cell and Ti using the metalorganic precursor titanium tetraisopropoxide (Ti(OiPr)₄, TTIP), the relatively complex thermal decomposition process of TTIP has thus far prevented a detailed understanding of the growth kinetics at play.
In this talk, we will discuss the first steps towards resolving the mystery of the hybrid MBE growth process by studying the thermal decomposition of TTIP using reactive force field molecular dynamics (ReaxFF-MD) simulations.² We find that the total thermal decomposition takes place via multiple steps and is typically initiated by bond breaking between C-O bonds rather than Ti-O bonds without additional hydrogen abstraction, followed by multiple ligand bond breaking steps, all with hydrogen abstraction. Metadynamics simulations were in excellent agreement with the ReaxFF-MD, revealing that the energy barriers to break Ti-O bonds (237-451 kcal/mol) were 4-5 times higher compared to breaking C-O bonds instead (53-119 kcal/mol). Nevertheless, ligand bond breaking subsequent to the first C-O bond breakage can also involve Ti-O bonds, which is attributed to the over coordination of Ti with oxygen. Performing the ReaxFF-MD simulation in the presence of TiO₂ and SrO terminated SrTiO₃(001) surfaces revealed that the particular surface chemistry of SrTiO₃ growth front dramatically affected the initialization of the TTIP decomposition process. While the metalorganic was found to be rather stable with a pronounced tendency to thermally desorb from TiO₂ terminated SrTiO₃(001) surfaces, its decomposition on SrO terminated surfaces was much pronounced, providing first insights into the intricacies of the hybrid MBE growth process.

¹ M. Brahlek, A. Sen Gupta, J. Lapano, J. Roth, H. T. Zhang, L. Zhang, R. Haislmaier, R. Engel Herbert, Adv. Funct. Mater. 28, 1702772 (2018).
² T.P. Senftle, S. Hong, M.M. Islam, S.B. Kylasa, Y. Zheng, Y.K. Shin, C. Junkermeier, R. Engel Herbert, M.J. Janik, H.M. Aktulga, T. Verstraelen, A. Grama, A.C.T. Van Duin, npj Comp.Mater. 2, 15011 (2016).

Titanium nitride thin films can be fabricated by both thermal and Plasma-Enhanced Atomic Layer Deposition (ALD and PE-ALD, respectively). The prevalent precursor couple is TDMAT as Ti-source and NH₃ as co-reactant. This combination is suitable for both routes. Nevertheless, PE-ALD recipes found in the literature exhibit a slower growth rate compared to thermal ALD (see e.g. Ref [1]). This can be disadvantageous for industrial development. However, in addition to NH_3, PE-ALD offers a larger palette of nitriding agent such as N₂, H₂ and N₂/H₂. This work compares therefore the PE-ALD of TiN using NH₃ and N₂, both combined with Ar, to evidence their influence on the final film properties and to maximize the growth rate. Parameters impacting the growth are monitored by in situ characterizations (ellipsometry, OES) while ex situ techniques (TEM/SEM/AFM, XPS, XRD, 4-probes measurements) are used to assess the morphology, the composition, the crystalline structure and the electric properties of the films. A similar approach is applied to the thermal route taken as reference.
To optimize the initial PE-ALD recipe, a set of parameters are investigated: the pulse and purge durations as well as the dilution of the N-sources within Ar and the plasma power. Those parameters are adapted to limit recombination reactions of the generated reactive species between the remote plasma source and the substrate. A large Ar dilution of both N₂ and NH₃ limits the film growth, while a 1:1 gas ratio for N-source and Ar leads to the optimal values. The pulse duration is set at 2 and 5 s for NH₃- and N₂-based plasmas, respectively. The purge duration is also optimized to shorten the ALD cycle duration. The applied plasma power (from 50 to 300 W) has no significant impact on the growth per cycle (GPC) with NH₃ while, using N₂, the GPC is maximum at the highest power. This is consistent with the expected low reactivity of N₂ (inert without plasma activation) regarding to highly reactive NH₃. These measurements are correlated to the plasma signals observed by OES. This indicates that producing less active species enhances their transport by limiting their recombinations. The films grown from both N-sources have a similar roughness, composition and morphology. However, the conductivity and the GPC are superiors using NH₃-based plasma. In addition, this process exhibits a remarkable conformality on high aspect ratio Si substrate for a PE-ALD route. The N₂ plasma process presents an acceptable film quality, it can be considered as well since it is more affordable, and it uses a non-harmful gas.
[1] L. Assaud et al. Highly-Conformal TiN Thin Films Grown by Thermal and Plasma-Enhanced Atomic Layer Deposition, ECS J. Solid State Sci. Technol.,2014, 3, 7, 253–258.

We report uniform film growth of SiO₂ on a large surface area (8-inch wafer) using synchronous modulated flow draw ALD (SMFD-ALD) technique. This technique allows greater precursor utilization and shorter deposition cycles time by modulating a pressure gradient between the reaction space and the pumping system. SiO₂ thin films are grown using BDEAS, ozone, and MMH precursors, wherein MMH acts as a catalyst. The use of catalyst is very important in SiO₂ film growth because without catalyst, surface reactions are generally very slow and occur at exceptionally high temperature. This SiO₂ film will be used as a capping layer in Al₂O₃/TiO₂ based nanolaminate capacitors to reduce the leakage current thereby improving working potential. We have already been able to fabricate capacitors with multi-stacked Al₂O₃/TiO₂ nanolaminate (~150 nm) with 4 nm of Al₂O₃ capping layer deposited by SMFD-ALD technology on both rigid and flexible substrates. The stacking of thin layers of these oxides results in a dielectric constant value that is one or two orders of magnitude higher than the individual oxides used due to a process known as Maxwell Wagner relaxation. On a planar substrate, these capacitors have the specific capacitance of ~1.6 /cm² and breakdown voltage of 3 V (~2 x10⁵ V/cm) with 10 leakage current. To lower the leakage and increase the breakdown voltage, SiO₂ will be employed as a capping layer and new capacitors will be fabricated on both planar and nano-structured high surface area electrodes, and the results will be presented. Since ALD technique is proved to produce uniform and conformal coating on high aspect ratio structures, it allows us to coat these dielectric nanolaminates in high surface area electrodes to further increase the energy density. As we have a well-established process for ZnO nanorods growth, nanostructured capacitors will be fabricated by coating Al₂O₃/TiO₂/SiO₂ nanolaminate on ZnO nanorods.

Silicon carbide (SiC) is a remarkable semiconductor material for high-temperature and high-voltage applications. It is also used as a material for a focus ring in etching process of silicon wafers, due to its high plasma resistance. SiC is commonly fabricated via chemical vapor deposition (CVD) or physical vapor transport (PVT) methods, and CVD-SiC is widely used as a material for the focus ring to protect the electrostatic chuck and create a uniform plasma condition during the plasma etching process of a silicon wafer. It is well known that a defect typically found in SiC known as a micropipe degrades the material's quality. A micropipe density is an important characteristic in SiC quality through SiC development due to the problem of reduced electron mobility.

Currently, CVD-SiC is widely used because it is favorable for manufacturing SiC with a low micropipe density. However, CVD-SiC has an issue in that the unit cost is high. PVT-SiC, on the other hand, offers the advantage of a cheap production cost despite the difficulty of reducing micropipe density. While it has been reported that micropipes reduce electron mobility in SiC, no specific influence on plasma resistivity has been identified.
In this study, we investigated the etching rate of CVD-SiC with small grain size and PVT-SiC with large grain size when SiC crystals grown by PVT and CVD methods are etched by reactive ion etching (RIE) under severe etching circumstances. In addition, we compared the etching behavior of micropipes and other sections.

As a result, the etching behavior in the micropipe was revealed to differ from that at the grain boundary. CVD-SiC with a small grain size has a problem in that SiC particles are peeled off grain by grain during the etching process, affecting the Si wafer quality. In the case of PVT SiC, however, the etching resistance was found to be higher due to lower grain boundary density. We also observed that, despite its relatively high micropipe density, PVT-SiC could be used as a focus ring material due to its high plasma resistance. The lower unit cost of the SiC focus ring, which is required as a consumable in the semiconductor process, is expected to increase production efficiency.

There is a broad and increasing interest in chemically controlled surfaces. Potential applications span from biotechnological and medical to antifouling and adhesive coatings. Chemical vapor deposition (CVD) polymerization of substituted [2.2]paracyclophanes is a versatile method that can result in a wide spectrum of functionalized surface coatings [1-3]. Emerging research involves the design of topologically controlled surfaces using templated CVD polymerization. Controlled deposition into thin liquid crystalline films resulted in supported polymer nanofiber arrays where intrinsic properties of the LC phase dictate the lengths, shapes and diameters of the nanofibers [3]. Building upon these earlier efforts, we explored the LC-templated CVD polymerization of chiral [2.2]paracyclophanes for the synthesis of nanohelices with superhierarchical chirality. Intrinsic properties of the precursors such as the side group chemistry and enantiomeric excess critically influence the properties of the resulting enatiomorphic polymer coatings. A more detailed understanding of the underlying mechanisms contributing to the chirality transfer from molecular to meso- and macroscopic scales is emerging and will inform the controlled synthesis of chiral interfaces for biomedical, optical and electronic properties.

[1] H.Y. Chen, J. Lahann, Designable Biointerfaces Using Vapor-Based Reactive Polymers, Langmuir 2011, 27(1), 34–48.
[2] X. Deng, C. Friedmann, J. Lahann, Bio-Orthogonal “Double-Click” Chemistry Based on Multifunctional Coatings, Angewandte Chemie International Edition 2011, 50, 6522-6526.
[3] A. Ross, H. Durmaz, K. Cheng, X. Deng, Y. Liu, J. Oh, Z. Chen, J. Lahann, Selective and reversible binding of thiol-functionalized biomolecules on polymers prepared by chemical vapor deposition polymerization, Langmuir 2015, 31, 5123-5129.
[4] K.C.K. Cheng, M.A. Bedolla-Pantoja, Y.-K. Kim, J.V. Gregory, F. Xie, A. de France, C. Hussal, K. Sun, N.L. Abbott, J. Lahann, Templated Nanofiber Synthesis via Chemical Vapor Polymerization into Liquid Crystalline Films, Science 2018, 362, 804–808.

Filled carbon nanotubes (CNTs) possess unique physical properties which arise because of the combined effects of the host CNTs and the guest fillers. Until now, many inorganic and organic materials have been successfully encapsulated inside CNTs, however, there is still a lack of a reliable method to achieve a complete filling of CNT cores with transition metal sulfide nanowires. Unlike the conventional ex-situ filling techniques which primarily lead to a partial filling, an in-situ filling method can be used to achieve a continuous and complete filling of the CNT channels. We have developed for the first time, a simple and reliable in-situ method to synthesize the Ni₃S₂ nanowires filled CNTs (Ni₃S₂@CNTs) directly on different substrates such as silicon, nickel foam, and carbon cloth using nickel salt or nickel nanoparticles as the catalyst precursor in a chemical vapor deposition system. The samples have been characterized using SEM, TEM, EDX, Raman spectroscopy, XRD, UV-Vis, TGA, and FTIR. The electron microscopy results show a linear and a tapering structure of Ni₃S₂@CNT completely filled with a continuous and single-crystalline Ni₃S₂ nanowire up to several micrometers in length whereas the Raman spectroscopy indicates a highly graphitized multi-walled CNT system (I_D/I_G= 0.26). The intrinsic I-V characteristics of individual Ni₃S₂@CNTs were studied using both two probe and four-point probe methods which reveal the metallic properties of Ni₃S₂@CNTs with a mean resistivity of 6.1×10^-5 Ωm. Also, it was observed that an individual Ni₃S₂@CNT having a small diameter (below 150 nm) can undergo an electrical breakdown at a higher current causing the encapsulated Ni₃S₂ nanowire to extrude out from the breakdown region whereas the Ni₃S₂@CNT with a bigger diameter can be emptied by pushing out the Ni₃S₂ nanowire along the direction of the current. Given the high quality and metallic behavior of Ni₃S₂@CNTs, they can find potential applications in future nanoelectronics devices and systems.
This work is supported by the National Science Foundation under Grant DMR- 1506640.

High refractive index polymers (HRIP) are an incredibly important class of materials given their potential applications in optoelectronic devices, such as anti-reflective components for displays, encapsulants for light-emitting diodes, and optical sensors. Because typical commodity polymers possess refractive indexes (RI) in the range 1.3 – 1.7, obtaining polymers with RI greater than 1.8 while maintaining high optical transparency is an important challenge. Halogen atoms such as chlorine, bromine, and iodine have high atomic refractivity ranging from 5.97 to 13.90, making them useful components in the development of HRIPs.
In this work, a series of halogenated polymer thin films with high refractive indexes (n =1.7 - 2.0) and outstanding optical transparency in the visible range, were prepared via a unique vapor phase process – initiated chemical vapor deposition (iCVD). iCVD is a versatile technique that has been used to fabricate a wide range of polymer coatings for many applications. In this study, poly(4-vinylpyridine) (P4VP) films with thickness ranging from 100nm to 600nm were prepared using iCVD, then treated by halogen vapor to form halogen-containing polymer thin films with high RI. The charge-transfer complex (CTC) formed between P4VP and halogen compounds, prepared by a simple vapor phase infiltration of halogen compounds, increases the RI of the P4VP thin film from 1.58 to 2.0 or higher, while maintaining the conformal and smooth nature of as-deposited thin films. Specifically, the P4VP-I₂ complex is demonstrated to have an RI of 2.0 and is transparent above a wavelength of 600nm. In another formulation, P4VP complexed with ICl achieved an RI up to 1.77, while still retaining the outstanding optical transparency throughout visible range. The RI of the halogenated polymer films can be further fine-tuned by controlling the concentration of CTC in the polymer film via copolymerization with other monomers that are inert to halogen compounds. In this work, a series P4VP thin films copolymerized with ethylene glycol dimethacrylate (EGDMA) with RI ranging from 1.50 to 1.98 were prepared.
The formation of charge-transfer complex (CTC) in halogenated polymer thin film was confirmed via both Fourier-transform infrared and UV-Vis spectroscopies. The optical performance of these coatings was characterized by spectroscopic ellipsometer and specular reflectometry. The thermal resistance and environmental stability of these films were also investigated as a function of temperature. The refractive index (RI) of P4VP-I₂ film decreased by 7.9% after 24 hours at 20°C. In comparison, the RI of P4VP-ICl film decreased by 0.41% over 24 hours at 20°C. These halogenated polymer thin films are expected to have wide applications in sensors and optoelectronic devices.

Area-selective deposition (ASD) is a process that controls where the deposition takes place through the underlying surface rather than through any masking step. This process can be achieved by either Atomic Layer Deposition (ALD) or Chemical Vapor Deposition (CVD) techniques [1]. In previous work, we studied area selective plasma enhanced CVD or PECVD using Ar/SiF4/H2 plasma chemistry [2]. For specific plasma parameters, a microcrystalline silicon film is selectively grown on a SiOxNy area while the AlOx adjacent area remains pristine (see figure 1). This effect was then attributed to the formation of Al-F bonds that blocks the deposition of silicon on top of the AlOx area [3].
However, when the plasma conditions are slightly changed or when those two materials are patterned, the selectivity is lost. Each case will be discussed and presented based on in-situ ellipsometry, atomic force microscopy and x-ray photoelectron spectroscopy analyses.
[1] G. N. Parsons and R. D. Clark, Chem. Mater. 32, 4920 (2020).
[2] G. Akiki, D. Suchet, D. Daineka, S. Filonovich, P. Bulkin, and E. V. Johnson, Appl. Surf. Sci. 531, 147305 (2020).
[3] G. Akiki, M. Fregnaux, I. Florea, P. Bulkin D. Daineka, S. Filonovich, M. Bouttemy and E. V. Johnson, J. Vac. Sci. Technol. A 39, 013201 (2021)

Desalination is becoming a vital part of the solution to water scarcity with a current worldwide desalination capacity of approx. 100 million m³/day, which is expected to grow by 40% by 2030. In 2019, 29% of the global desalination capacity was of brackish and river water (500-20000 ppm TDS). Most (98%) of the brackish water nowadays is desalinated by reverse osmosis (RO) which completely deprives the permeate of Mg²⁺ and Ca²⁺. Both are critical nutrients needed to sustain human health and agriculture. Replenishing these divalent ions to the desalinated water increases the cost and complexity of the overall process and decreases its sustainability and economic attractiveness. Electrodialysis is energetically more efficient than RO for desalination of brackish water and using monovalent selective membranes—membranes that enable monovalent ions passage, but retain divalent ions—offer a potential solution to preserve the nutrients in the permeate. However, current methods for instilling ion selectivity use large macromolecular building blocks to form the selective layer, which increases ion transport resistance and hence the energy demand and cost of selective electrodialysis. To disrupt the resistance-selectivity trade-off, a method for uniformly depositing an ultrathin and highly charged selective layer is needed.
While low-temperature atomic layer deposition is a suitable method for depositing ultrathin metal oxides on membranes, the elastic mismatch between metal oxides and polymers hinders the stability of the selective layer. Instead, we used molecular layer deposition of alucones to resolve the elastic mismatch. We show that the electrostatic charge of alucone surface leads to monovalent selectivity values on par with commercial membranes, but with only a marginal increase in resistivity. We further investigate the resistance-selectivity trade-off using electrodialysis simulation and suggest “guidelines” for further improvement of performance.

Nanomaterials, in which nanoparticle size-dependent properties are maintained, but also new collective properties derived from the synergy between their individual building blocks emerge, gained significant interest in recent decades. However, the lack on structural integrity when the dimensions are upscaled still limits them to microscale applications. The lack of control over larger length scales when nanomaterials are processed usually leads to defects which scale with the dimensions of the specimen. A promising strategy to overcome this issue is inspired by nature. Nature in fact evolved hard, strong, and tough materials-like bone, teeth, and nacre- by hierarchically structuring nano building blocks with soft interfaces on different length scales.

The synthetic emulation of these multiscale, hierarchical structures based on hard minerals combined with soft biopolymers with structural features down to the nanoscale bears tough practical challenges on each length scale. These challenges start at the atomic level, and expand up to the final macroscopic material and its application. Therefore the cooperation from different scientific disciplines ranging from chemistry and physics to material science and engineering becomes crucial.
This talk will discuss how these different hurdles can be addressed via a bottom-up approach using organo-functionalized ceramic nanoparticles. It will be shown how the quest begins on the atomistic scale, aiming to boost and control the materials’ strength by a crosslinking reaction of the nano-building blocks as well as their self-assembly at the nanoscale level. At the microscale, it will be shown how emulsion-based approaches allow to control the shape of nano- to micro-architected suprastructures. These ceramic nanocomposite suprastructures can be afterwards functionalized in order to process them on the mesoscale with polymers ultimately leading to robust macroscale materials in which nano- and micro-features are maintained.

Flexible optics and optoelectronics are becoming increasingly desirable in many advanced technologies. One example is varifocal liquid lenses mimicking the mechanism of a human eye, where transparent fluid is encapsulated between two flexible membranes. Varifocal liquid lens is particularly useful in volume-limited applications, such as miniature medical devices, smartphone cameras, and wearable augmented reality equipment. The focal length is modulated by mechanically actuating the membrane–similar to the ciliary muscle in the human eye. These optical devices present significant material design challenges; flexible, transparent coatings are required to inhibit the diffusion of high-index fluids (optical oil) into the elastomeric membrane, which would otherwise degrade the optical quality of the lens. Conventional optical coatings based on inorganic compounds, such as silica and alumina, can provide the required barrier performance but their intrinsic brittleness limit compatibility with polymer-based flexible optics/optoelectronics. Herein, a flexible and transparent polymer barrier layer was created by initiated chemical vapor deposition (iCVD) of a highly crosslinked and fluorinated diacrylate–poly(1H,1H,6H,6H-perfluorohexyl diacrylate) (pPFHDA). The deposition was realized by a bubbler-enhanced iCVD which enables vapor delivery of nonvolatile monomers such as PFHDA. During the deposition, the monomer and the substrate were kept at around room temperature, and the deposited coating showed high uniformity and conformality. This was believed to be the first report of vapor deposition of pPFHDA. The pPFHDA coating has no optical absorption from the wavelength of 300 nm to 1690 nm and is thermally stable to 300°C. Coatings with thicknesses in hundreds of nanometers were found to be impermeable to high-index optical fluid for more than two months at 70°C, which is equivalent to a four-year lifetime at room temperature. Such barrier performance was achieved even after the layer was subject to equi-biaxial strain, which demonstrated that the pPFHDA barrier layer was able to accommodate mechanical deformation, making it a compelling solution for flexible optics/optoelectronics.

Spatial Atomic Layer Deposition (SALD) is a recent variant of ALD that offers fast processing, even at atmospheric pressure, while preserving the unique assets of ALD, namely, precise thickness control down to the nanometer, high-quality films even at low temperatures, and unique conformality. As a result, SALD is ideal for applications requiring high throughput at low cost, such as new generation photovoltaics, LEDs or packaging. In particular, the SALD approach based on close-proximity deposition heads is highly versatile since it can be easily customized by proper design of the deposition heads and because the deposition takes place in the open air without the need of any deposition chamber.

In my talk, I will first present the developments on close proximity SALD we are currently carrying out at LMGP. In particular, I will illustrate how 3D printing can be used to prototype and customize the deposition heads, and how in so doing SALD can indeed be tuned to deposit custom patterns without the need of pre-patterning steps. This approach represents a new versatile way of printing functional materials and devices with spatial and topological control, thus extending the potential of SALD, and of ALD in general.
I will then present recent studies showing the effect of air-processing on the properties of the thin films deposited with our close-proximity system, and how the choice of precursor can have a huge impact on the final properties of the materials deposited.
The potential of SALD will be illustrated through different examples of applications of the thin films developed in our group.

In this work we report a new area-selective atomic layer deposition (AS-ALD) approach enabled by a spatially controlled hydroxylation process. The processing occurs within a low-pressure water vapor ambient established within an environmental scanning electron microscope (ESEM). The ESEM electron beam interacts with the water vapor and generates a local region of reactive species (e.g. hydroxyl radicals) in the neighborhood of the focused electron beam. Here, we functionalize exfoliated graphene substrates with a surface hydroxyl layer by selectively scanning the electron beam within the region to be patterned. The electron-beam generates reactive hydroxyls that then dock on the graphene surface. We emphasize that graphene is natively resistant to surface hydroxylation using conventional exposure to water vapor at elevated temperature. As a result, conventional ALD deposition on graphene is limited to defects and edges where functional groups may take root. Our area-selective hydroxylation technique produces dense and well-patterned hydroxylation on a graphene surface, resulting in high-quality patterned ALD deposition.

In the current study, hydroxyl functionalization and ALD deposition occur within 2 x 2 micron patterned regions on the surface of graphene flakes exfoliated from HOPG. A nominal 10 nm layer of alumina and zinc oxide is then deposited using conventional ALD methods. The hydroxylated regions are sufficiently stable to withstand the ALD deposition temperature of 200 ^oC. We show that the hydroxyl functionalization and resulting ALD deposition efficacy is highly dependent on electron beam dwell time, electron beam current, pressure, and electron beam acceleration voltage. These processing factors have previously been shown to direct the radiolysis reaction products formed during electron collisions with water vapor. Importantly, all successful hydroxylation experiments required an electron beam dwell time that far exceeds conventional SEM imaging parameters. The patterned hydroxyl functionalization and resulting ALD coatings are characterized using scanning electron microscopy, energy dispersive spectroscopy (EDS) mapping, atomic layer deposition, and Raman spectroscopy. We foresee that this technique may be used to deposit high-quality oxides to traditionally ALD-resistant materials to form field-effect transistor devices, metamaterials, and other nanoscale or microscale devices.

Although carbon nanotubes have a wide variety of uses in electronics, structural composites, and energy storage, questions remain regarding specific growth mechanisms. The interactions between the carbon precursor, the metal catalyst, and the substrate are complex and involve the evolution of the metal catalyst oxidation state under hydrogen and hydrocarbon reductants. As carbon nanotube growth yield is often less than 10%, understanding the growth mechanism may provide insights for more efficient synthesis. We study chemical vapor deposition of carbon species in-situ in the transmission electron microscope (TEM) at atmospheric pressure in order to better understand the parameters that most strongly influence the morphology of carbon grown. Recent studies have shown that Na-based catalysts were observed to exhibit significantly reduced growth temperatures for carbon nanotubes. Here we report the influence of alkali metal salts on Fe and Ni based catalysts for carbon fiber growth. In-situ chemical vapor deposition experiments in the TEM with ethylene hydrocarbon precursors showed that Na₂CO₃ led to significantly increased yield in carbon nanofiber growth, which we hypothesize to be related to a desiccating effect. We combine these in-situ growth experiments in the TEM with ex-situ growth experiments on the influence of Na₂CO₃ additive on ex-situ carbon nanofiber growth using various substrates and carbon precursors. In-situ chemical vapor deposition in the TEM is promising for insights into the growth mechanism for carbon based materials, which exhibit a wide range of structures.

Vapor phase infiltration (VPI) infuses the bulk of polymers with vapor phase metalorganic precursors which are often co-reacted with an oxidant to form air stable hybrid materials. Due to its vapor phase nature, the VPI process can be utilized as for post-fabrication modification of a wide range of polymeric forms (thin films, fabrics, etc.) without significantly modifying the material’s macroscale shape. To date, VPI has shown promise in several industrially relevant areas including: the creation of solvent stable hybrid membranes for chemical separations, the modification and stabilization of mechanical properties for polymer and biomaterial-based fibers, and the creation of nanoscale metal oxide structures from block copolymer templates. The properties of these hybrid materials have shown significant dependence on both their chemical and physical structure. This structure is determined by inherent properties of the polymer and precursor system in addition to the processing parameters employed during VPI. Developing a complete understanding of the process-structure-property relationships for hybrids created via VPI is necessary for specific application design and scale-up. In this talk, we will understand the temperature dependent property of solvent stability by exploring changes in the chemical and physical structure of AlO_x / poly(methyl methacrylate) [PMMA] hybrid materials. In situ quartz crystal microbalance gravimetry (QCM) will be leveraged to study the long-term mass change behavior of precursor sorption and desorption. Specifically, this work explores the role of temperature on the total mass sorption of TMA into PMMA with insight into contributions from the free diffusing and chemically interacting species. Consistent with literature reports, we find the total precursor uptake for this VPI system displays a complex dependency on temperature with an increase in quantity from 70 ^oC to 90 ^oC followed by a continuous decrease with further increasing temperature. Using QCM experimentation, we determine that the initial increase is driven by reaction kinetics where the reacted quantity increases with temperature and then saturates at approximately 110 ^oC. However, we find the quantity of free diffusing species decreases continuously with temperature and drives the decrease in uptake at higher temperatures. This information is used to explain quantitatively the temperature dependent property of solvent stability observed in this VPI system as well as provide a framework for studies of this nature. Additionally, we extract kinetic behaviors of this VPI system as a function of infiltration temperature for use in scaling the VPI process to industrial levels.

Solid layers^1,2 of ZrO₂ are often deposited by chemical vapor deposition (CVD), where a metal-containing precursor is evaporated and forms a film in a surface reaction. To design such a process, the gas phase reactions, especially their temperature-dependent kinetics, should be known, which is seldom the case. Intermediates, formed in the gas phase can lead to film formation or to a depletion of the precursor and therefore a reduction in growth rate. The analysis of these early stages of growth requires fast and sensitive analytical techniques with sufficiently low detection limits for elusive gas phase species. Experimental limitations precluded the detection of most of the postulated intermediate species, and reaction mechanisms are predominantly hypothetical so far.^3,4 An experimental approach to enlighten the gas phase chemistry of an often used precursor, zirconium acetylacetonate, is presented here, using a microreactor coupled to a mass spectrometer with mild ionization from a synchrotron source. This was demonstrated previously, to be an effective way to detect and characterize gas phase intermediates with short lifetime.^5,6
A combined numerical and experimental study of the vacuum pyrolysis of zirconium acetylacetonate, Zr(C₅H₇O₂)₄, in a microreactor at short residence times of < 60 μs was carried out. The precursor is sublimed, subsequently transported by argon or helium carrier gas, and expanded through a pinhole into a resistively heated 1 mm inner diameter SiC-microreactor. Products leaving the reactor are expanded and ionized by tuneable vacuum ultraviolet (VUV) synchrotron radiation and characterized by imaging photoelectron photoion coincidence spectroscopy (i²PEPICO) and mass spectrometry at the Swiss Light Source in Switzerland. From the temperature-dependent photoionization mass spectra between 130 and 650 °C, two initial decomposition steps were identified and temperature regimes for the following pathways of thermal decomposition at higher reactor temperatures were determined. In addition, we recorded photoionization efficiency curves and threshold photoelectron spectra (TPES) at photon energies of 6.5-11.5 eV, which gave us direct evidence about the intermediates and products formed. Six important Zr-intermediates were identified for the first time, as for example Zr(C₅H₇O₂)₂(C₅H₆O₂), and ZrO(C₅H₇O₂)₂, together with several organic products, which aided us to describe the predominant thermolysis pathways in the temperature range of 130-650 °C. In conjunction with the simultaneous numerical simulation of the flow field in the microreactor, the application of synchrotron radiation coupled to the i²PEPICO experimental apparatus is a promising way to enlighten pyrolysis pathways and kinetics of metal-organic precursors.


References
1. A. Méndez-López, O. Zelaya-Ángel, M. Toledano-Ayala, I. Torres-Pacheco, J.F. Pérez-Robles and Y.J. Acosta-Silva, Crystals, 2020, 10(6), 454.
2. J.G. Thakare, C. Pandey, M.M. Mahapatra and R.S. Mulik, Met. Mater. Int., 2020, 366, 338.
3. M.V.F. Schlupp, J. Martynczuk, M. Prestat and L.J. Gauckler, Adv. Energy Mater., 2013, 3(3), 375.
4. H.M. Ismail, Powder Technology, 1995, 85(3), 253.
5. Z. Zhang, Y. Pan, J. Yang, Z. Jiang and H. Fang, Journal of Crystal Growth, 2017, 473(12), 6.
6. P. Hemberger, J.A. van Bokhoven, J. Pérez-Ramírez and A. Bodi, Catal. Sci. Technol., 2020, 10(7), 1975.

Although carbon nanotubes have a wide variety of uses in electronics, structural composites, and energy storage, questions remain regarding specific growth mechanisms. The interactions between the carbon precursor, the metal catalyst, and the substrate are complex and involve the evolution of the metal catalyst oxidation state under hydrogen and hydrocarbon reductants. As carbon nanotube growth yield is often less than 10%, understanding the growth mechanism may provide insights for more efficient synthesis. We study chemical vapor deposition of carbon species in-situ in the transmission electron microscope (TEM) at atmospheric pressure in order to better understand the parameters that most strongly influence the morphology of carbon grown. Recent studies have shown that Na-based catalysts were observed to exhibit significantly reduced growth temperatures for carbon nanotubes. Here we report the influence of alkali metal salts on Fe and Ni based catalysts for carbon fiber growth. In-situ chemical vapor deposition experiments in the TEM with ethylene hydrocarbon precursors showed that Na₂CO₃ led to significantly increased yield in carbon nanofiber growth, which we hypothesize to be related to a desiccating effect. We combine these in-situ growth experiments in the TEM with ex-situ growth experiments on the influence of Na₂CO₃ additive on ex-situ carbon nanofiber growth using various substrates and carbon precursors. In-situ chemical vapor deposition in the TEM is promising for insights into the growth mechanism for carbon based materials, which exhibit a wide range of structures.

Vapor phase infiltration (VPI) infuses the bulk of polymers with vapor phase metalorganic precursors which are often co-reacted with an oxidant to form air stable hybrid materials. Due to its vapor phase nature, the VPI process can be utilized as for post-fabrication modification of a wide range of polymeric forms (thin films, fabrics, etc.) without significantly modifying the material’s macroscale shape. To date, VPI has shown promise in several industrially relevant areas including: the creation of solvent stable hybrid membranes for chemical separations, the modification and stabilization of mechanical properties for polymer and biomaterial-based fibers, and the creation of nanoscale metal oxide structures from block copolymer templates. The properties of these hybrid materials have shown significant dependence on both their chemical and physical structure. This structure is determined by inherent properties of the polymer and precursor system in addition to the processing parameters employed during VPI. Developing a complete understanding of the process-structure-property relationships for hybrids created via VPI is necessary for specific application design and scale-up. In this talk, we will understand the temperature dependent property of solvent stability by exploring changes in the chemical and physical structure of AlO_x / poly(methyl methacrylate) [PMMA] hybrid materials. In situ quartz crystal microbalance gravimetry (QCM) will be leveraged to study the long-term mass change behavior of precursor sorption and desorption. Specifically, this work explores the role of temperature on the total mass sorption of TMA into PMMA with insight into contributions from the free diffusing and chemically interacting species. Consistent with literature reports, we find the total precursor uptake for this VPI system displays a complex dependency on temperature with an increase in quantity from 70 ^oC to 90 ^oC followed by a continuous decrease with further increasing temperature. Using QCM experimentation, we determine that the initial increase is driven by reaction kinetics where the reacted quantity increases with temperature and then saturates at approximately 110 ^oC. However, we find the quantity of free diffusing species decreases continuously with temperature and drives the decrease in uptake at higher temperatures. This information is used to explain quantitatively the temperature dependent property of solvent stability observed in this VPI system as well as provide a framework for studies of this nature. Additionally, we extract kinetic behaviors of this VPI system as a function of infiltration temperature for use in scaling the VPI process to industrial levels.

Cuprous oxide (Cu2O) is a promising material for emerging electronics. Bandgap and defect states estimation of Cu2O film grown by chemical vapor deposition (CVD) technique on SiO2/Si substrate at 600°C was performed in this work using various spectroscopies. XRD and XPS confirmed the phase purity of deposited film. The optical bandgap of Cu2O was 2.35 eV which was calculated from absorption spectroscopy (UV-Vis). Using ultraviolet photoelectron spectroscopy (UPS) and absorption spectroscopy (UV-Vis), the nature of deposited Cu2O was confirmed as a p-type semiconductor. Using UPS, the energy level of the valence band and Fermi level of the intrinsically p-doped Cu2O film were determined as 6.02 eV and 5.62 eV respectively from the vacuum level. The temperature dependence photoluminescence spectra of these films exhibited emission peaks at ~1.4 eV and ~1.72 eV which corresponds to energy states of copper and oxygen vacancies level in the bandgap. Photoluminescence spectra also show a peak at ~2 eV which is due to the relaxation of free excitons to the valance band.

The ALD/MLD (atomic/molecular layer deposition) technique allows us to fabricate novel hybrid materials not readily accessible through any other fabrication route. These include exciting in-situ crystalline metal-organic framework (MOF) like materials and inorganic-organic interfaces and superlattice (SL) structures in which ultra-thin organic layers are introduced periodically between nm-scale metal oxide layers to e.g. enhance mechanical flexibility, block phonon conduction, or bring photoactivity.^[1,2] In this presentation I will briefly discuss exciting examples of properties/functionalities realized for these materials: (i) electroactive lithium-organic thin films for battery components (electrodes, solid electrolytes) and artificial solid electrolyte interphase (SEI) layers;^[3,4] (ii) lanthanide-organic upconverting layers for solar cells;^[5] (iii) ZnO-based oxide:organic SL films for textile-integrated thermoelectrics;^[6] (iv) ε-Fe₂O₃:azobenzene SL films for flexible and photo-responsive room-temperature magnets.^[7]

References
[1] A. Khayyami, A. Philip, M. Karppinen (2019) 13400.
[2] R. Ghiyasi, M. Milich, J. Tomko, P.E. Hopkins, M. Karppinen, Appl. Phys. Lett. 118 (2021) 211903.
[3] J. Multia, J. Heiska, A. Khayyami, M. Karppinen, ACS Appl. Mater. Interfaces 12 (2020) 41557.
[4] J. Heiska, M. Madadi, M. Karppinen, Nanoscale Adv. 2 (2020) 2441.
[5] A. Ghazy, M. Safdar, M. Lastusaari, A. Aho, A. Tukiainen, H. Savin, M. Guina, M. Karppinen, Sol. Ener. Mater. Sol. Cells 219, 110787 (2021).
[6] G. Marin, R. Funahashi, M. Karppinen, Adv. Eng. Mater. 22 (2020) 2000535.
[7] A. Philip, J.-P. Niemelä, G.C. Tewari, B. Putz, T.E.J. Edwards, M. Itoh, I. Utke, M. Karppinen, ACS Appl. Mater. Interfaces 12 (2020) 21912.

The growth of complex oxide thin films with atomic precision offers bright prospects to study new physics emerging from the films and at their interfaces. Unfortunately, the ALD synthesis of a ternary oxides is, in most of the cases, not a simple addition of the corresponding binary oxide processes. The compatibility of different chemical precursors confined under similar deposition conditions together with a varying surface species after every precursor dose increase considerably the level of intricacy. As a case example, here we will present the challenging synthesis of Gd_xFe_yO_z phase by combining two homometallic Gd and Fe compounds using an ALD-type approach. We investigate the use of tailor-made metalorganic precursors, designed to provide similar thermal behavior, (bis(N-isopropyl ketoiminate) iron(II), [Fe(ⁱpki)₂] and tris(N,N′-diisopropyl-2-dimethylamido-guanidinato)gadolinium(III), [Gd(DPDMG)₃]). The deposited films resulted in cation ratio close to nominal stoichiometry with negligible amount of organic species according to the RBS/NRA analysis. Upon thermal treatment, XRD and SQUID measurements confirmed the formation of the garnet Gd₃Fe₅O₁₂ phase coexisting with GdFeO₃, Gd₂O₃ and Fe₂O₃. [1]
Additionally, when epitaxial growth is pursued for enhanced physical properties, the growth is further limited to specific fragile and expensive substrates hindering the opportunity to integrate such materials into polymeric and silicon substrates. To broaden the applicability of these materials we went one step further developing an all-chemical deposited route based on the use of a water-soluble sacrificial layer, to manipulate complex crystalline oxides as free-standing membranes and integrate them on polymer substrates such as PDMS and PET. [2] As a case example, we will present the synthesis of ALD-CoFe₂O₄ membranes followed by the structural and magnetic characterization by means of STEM and SQUID, respectively.
This is a robust chemical and low-cost methodology that could be adopted to prepare a wide variety of complex oxide films to fabricate artificial heterostructures to go beyond the traditional electronic, spintronic, and energy storage and conversion devices.

[1] P. Yu, M.Coll et al. CrystEngComm, 2021,23, 730-740
[2] P. Salles, M. Coll et al. Adv. Mater. Interfaces 2021, 8, 2001643

Although n-type amorphous metal oxide semiconductors have received commercial traction, and particularly amorphous indium gallium zinc oxide (a-IGZO), large-area electronics is still largely dominated by silicon-based technologies. The advance and development of both display and solar cell technologies is limited by the relatively poor performance or high temperature processing conditions of amorphous silicon or polycrystalline silicon devices. The commercial success of metal oxide semiconductors would be significantly enhanced if there was a good p-type metal oxide material which could complement a-IGZO in low power circuits based on thin film transistors (TFTs) and p-n junctions as required by the display and solar cell industries respectively.

Cuprous oxide (Cu₂O) is a non-toxic, low cost metal oxide that exists naturally as a p-type semiconductor due to intrinsic copper vacancies [1]. Reported Hall mobilities of ~10² cm²V^-1s^-1 suggest potential as a good thin film electronic material [2], but with sputter-deposited Cu₂O TFT field-effect mobilities of ~10^-1 cm²V^-1s^-1 and high off-state currents widely reported [3], further research is required to optimise the deposition and device fabrication.

This work looks at the deposition of Cu₂O films using plasma enhanced chemical vapor deposition (PECVD) with pulsed liquid injection of metal-organic precursors. PECVD is used for large-area amorphous silicon growth and has the advantage over physical vapor deposition techniques, such as sputtering, of allowing chemical control of the deposition process and hence potentially the oxidation state of the copper oxide thin film. The pulsed liquid injection is achieved with a Kemstream Vapbox liquid precursor delivery system and Cupraselect (Cu(hfac)(TMVS)) is used as the copper precursor with oxygen as the oxidising agent.

Existing literature using Cupraselect is primarily focussed on CVD of copper films and the presence of a significant incubation period is widely reported, with film deposition delayed by as much as 50 minutes depending on process conditions [4]. However, using PECVD and pre-treating the substrates with oxygen plasma, this work reports the deposition of 20nm thick Cu₂O films from a single 1 ms injection of precursor. Such film thicknesses would be ideal for deployment in TFT structures.

References
[1] H. Raebiger, S. Lany and A. Zunger, Phys. Rev. B, 76, 2007.
[2] B. S. Li, K. Akimoto and A. Shen, J. Cryst. Growth, 311, 2009.
[3] S. Han, et al., Appl. Phys. Lett., 109, 2016.
[4] S. Kim, J. M. Park, and D. J. Choi, Thin Solid Films, 315, 1998.

Acknowledgement - The authors thank EPSRC for providing the funding for this work through grant numbers EP/M013650/1, EP/P027032/1 and the EPSRC Centre for Doctoral Training in Integrated Photonic and Electronic Systems.

There are many fields within the physical and engineering sciences, where the key to scientific and technological progress is the understanding and control of the properties of matter at the scale of individual atoms and molecules, so conducting a study of the mechanisms that occur within the processes that allow us to manipulate matter on this scale; which has become a fundamental part of scientific development in its different areas. However, to carry out the study of nucleation and growth of matter at the atomic-molecular scale few techniques allow us to carry out the growth of materials with such precise control, and in techniques that are capable of doing it, there is a great ignorance of all the phenomena and interactions that occur in them due to the complexity involved in the study at this scale. ALD is one of the few techniques that provide control of film growth at the molecular atomic scale and a high precision of synthesis of nanocomposites, which makes ALD the ideal method to carry out the study of nucleation and growth of the different nanocomposite. Where one of the approaches of greatest interest within the fielid tess is the area of functional hybrid materials at the nanoscale since it is one of the most promising and emerging research areas in materials chemistry. This is because the limitless possible combinations of the different properties of inorganic, organic, or even bioactive components in a single material, in nanometric dimensions, have attracted considerable attention. This approach provides the opportunity to create a host of novel and advanced materials with well-controlled structures and multiple functions. Where the TiO2 / graphene hybrid stands out, which is a new group of nanomaterials whose use and performance are tested with greater thanks to its improved properties and possible applications, which, compared to pure materials, this composite material exhibits improved activities due to the synergistic effect of the GR and TiO2 components. This material is widely used in the area of thin films, so the ideal technique for its growth is ALD thanks to the properties it has and allows the TiO2 / graphene hybrid to grow in a very controlled and precise way, but most of the information on the growth mechanisms is very little known, and to have a greater efficiency of the technique and the materials, it is very important to know all the factors that involve the growth of the material. However thanks to the self-limiting nature of the chemisorption reactions during the ALD growth of the film. allows us to carry out a theoretical study through calculations using density functional theory (DFT) to develop a model of the chemistry involved in the reactions that occur in the ALD process during the early stages of growth of TiO2 on graphene for each grown film. Theories of the electronic structure of matter have become a very useful and powerful tool for understanding the chemistry involved in the early stages of the ALD process. In this work, we have used DFT calculations to present a complete reaction mechanism between OH-functionalized graphene and the 3 main Ti precursors (TDMAT, TIP, TiCl4) using water as the oxidizing gas. We have studied the ligand exchange reactions. A theoretical-computational study of the initial stages of ALD growth of the TiO2-GR hybrid was carried out from graphene functionalized with hydroxyl (OH) groups as growth sites for the interaction with the main titanium precursors in ALD. It was demonstrated how the chemical nature of the precursor directly affects the reaction mechanism to be carried out, reflecting on the energy required to invest in the system to overcome the barriers and for the reaction to take place. Additionally, we obtained the secondary products that the system produced when carrying out the growth of the film, with which a reaction mechanism was proposed for each ALD half cycle.

Two-dimensional (2D) semiconductors such as monolayer MoS₂ are essential building blocks for next-generation ultrathin flexible and low-power electronics. A recent study using semimetal Bi as the electrical contact to monolayer MoS₂ has improved the on-current and contact resistance of the monolayer transistor to be on par with traditional Si-based transistors. As a premise for large-scale electronics, batch production of the 2D semiconductors requires large domain size, large-area continuity, and thickness uniformity. These are difficult tasks in particular for synthetic MoS₂ and other transition metal dichalcogenides (TMDs), considering the less controllable mass flux from solid metal precursors in chemical vapor deposition (CVD), as compared to the case of graphene growth with gaseous hydrocarbons. One recent advance to mitigate this is the use of alkali metal halide salts [e.g., NaCl, KI, and NaBr], which, in conjunction with transition metal or metal oxide powders, could increase the mass flux of metal precursors and accelerate the 2D growth of TMDs. Various research efforts have been made to explain the mechanism, in terms of these metal halides as a molten salt to facilitate the evaporation of the metal oxide precursor, or as a surfactant to modify the substrate surface and the MoS₂ edge. Both the alkali metal ions and the halogens have been proposed to play an important role in the promoted growth. Continued endeavors are being devoted to bringing further lights in understanding the detailed mechanism.
In this work, we provide unambiguous experimental evidence that halogens are closely related to MoS₂ growth dynamics. To achieve this, post-growth Arrhenius analysis is implemented without any involvement of in-situ characterizations. We find that within the same reaction family, the halide-assisted growths conform to Brønsted-Evans-Polanyi (BEP) relation, where their reaction barriers are linearly correlated to the Mo-X (X = I, Br, Cl, F, and O) bond dissociation energies (E_b), suggesting the substitution of Mo-X bonds by the Mo-S bonds to be the rate-limiting step for the CVD growth. Based on this, we propose a theoretical growth model that not only reproduces the BEP relation but also explains the sulfur concentration-dependent growth dynamics observed in our experiments. By harnessing the synergistic effect of the KI promoter and the sulfur supply, we can reproducibly and rapidly synthesize near millimeter-sized 2D MoS₂ crystals dispersed over the entire SiO₂/Si substrates. These results not only shed lights on the detailed mechanism of TMD growth activated by the halide salts, but also guide the designer growth towards larger domain sizes that should enable practical applications.

Functional and responsive surfaces have been successfully deposited by initiated Chemical Vapor Deposition (iCVD) on a variety of substrates. Such method allows the full retention of the functional groups of the monomer structure, including, but not limited to, C-/N- containing ones. The high versatility of iCVD in driving application-specific properties into the material, creating a platform for the implementation of polymeric coatings into device fabrication, will be discussed.

A case study will be presented in the field of stimuli-responsive materials, in particular, hydrogels whose size and shape change when stimulated by light, humidity, temperature and aqueous environments. iCVD allows to obtain stimuli-responsive thin films with high chemical specificity and this is important to obtain a large responsiveness amplitude. In addition, the thin film form allows obtaining fast response.

Fast response and large signal amplitude are fundamental requirements for good sensors. Fast and ultra-fast humidity sensors based on the optical detection of the change in thickness of the iCVD hydrogels will be shown. The implemented sensor prototype delivered reproducible relative humidity values and the achieved response time for an abrupt change of the humidity was about three times faster compared to one of the fastest commercially available sensors on the market.

This presentation will focus on the applications of stimuli-responsive nanofilms for drug delivery devices, sensors and actuators. Their combination with piezoelectric/piezoresistive materials will also be discussed for innovative device geometries.

Two-dimensional materials (2DM), such as graphene and transition metal dichalcogenides (TMDs), always achieve great interest for future applications in electronics, optoelectronics and sensing. The atomic layer deposition (ALD) of ultra-thin high-κ dielectrics (among which Al₂O₃ and HfO₂) on the 2DM materials is an essential requirement to realize some high performance electronic devices, including lateral field effect transistor (FETs) and/or hot electron transistors (HETs). However, the lack of out-of-plane bonds in the 2DM makes that seeding-layers and/or pre-functionalization treatments are necessary to promote the ALD growth, despite the possible adverse impacts on the carrier mobility and the dielectric quality. In some special cases, the peculiar interaction between 2DM and the underlying substrate proved to be beneficial to achieve a direct ALD growth, such as in the case of monolayer (1L) CVD graphene residing on catalytic-metals (Cu, Ni-Au) [1,2], or 1L epitaxial graphene on 4H-SiC [3,4]. So far, similar substrate effects have not been explored in the ALD deposition of dielectrics on TMDs.
In this work, large area (in the order of cm) and high quality monolayer-MoS₂ were exfoliated from the bulk molybdenite on a gold layer and successively transferred onto Al₂O₃/Si substrate [5]. Hence, the direct ALD growth of ultra-thin (about 4 nm) Al₂O₃ and HfO₂ films was investigated on MoS₂ residing on these two different substrates. Employing the same deposition conditions, an optimal ALD-coverage has been obtained for monolayer-MoS₂ on gold (see Fig.1(a) for 80 cycles of Al₂O₃ deposition), while a highly inhomogeneous coverage was observed for monolayer-MoS₂ on the Al₂O₃/Si substrate (see Fig.1(b)). The beneficial effect of the gold substrate on the ALD gorwth has been correlated to the strong van-der-Waals interaction between the MoS₂ and Au, resulting in a tensile strain and p-type doping of the MoS₂ membrane, as shown by Raman mapping [5]. Good insulating properties of the compact ~4 nm thick Al₂O₃ membrane on Au-supported monolayer- MoS₂ have been found by conductive atomic force microscopy.
These results can have an important impact in the realization of devices based on large area MoS₂ membranes produced by gold assisted exfoliation.
This work has been supported by the FLAG-ERA JTC2019 project “ETMOS”
[1] B. Dlubak, et al., Appl. Phys. Lett. 100, 173113 (2012).

[2] A. I. Aria, et al., ACS Appl. Mater. Interfaces 8, 30564 (2016).

[3] E. Schilirò, R. Lo Nigro, S.E. Panasci, F.M. Gelardi, S. Agnello, R. Yakimova, F. Roccaforte, F. Giannazzo, Carbon 169, 172-181 (2020).

[4] E. Schilirò, R. Lo Nigro, F. Roccaforte, I. Deretzis, A. La Magna, A. Armano, S. Agnello, B. Pecz, I.G. Ivanov, R. Yakimova, F. Giannazzo, Adv. Mater. Interfaces 1900097 (2019).
[5] S.E. Panasci, E. Schilirò, G. Greco, M. Cannas, F. M. Gelardi, S. Agnello, F. Roccaforte, F. Giannazzo, ACS Applied Materials & Interfaces (2021). DOI: acsami.1c05185.

The technique of initiated chemical vapor deposition (iCVD) is gaining increasing interest in the fabrication of various polymer thin films. By implementing the chemistry of radical polymerization under the condition of vapor-based deposition, iCVD provides fine control of coating composition and functionality as well as adaptability and conformality to complex substrates. While the uniformity of polymer thin films is pursued in many fabrications, thin films with heterogeneous composition and morphology can provide high level of versatility, especially when holistic solutions are desired. We report our exploration of heterogeneous film fabrication using hybrid chemical vapor deposition with multiple film growth steps in one deposition. Besides changing process variables (such as vapor pressure and temperature) during iCVD, we introduced other fabrication procedures in situ to create polymer films with multi-scale heterogeneousness in vertical and planar dimensions. We will discuss three hybrid deposition process and the composition, structure, interfacial strength, surface property, and biocompatibility of the resultant thin films. The technical challenges encountered in the hybrid deposition process will also be discussed. We will conclude by demonstrating our recent work of implanted drug delivery, smart textiles, and sustainable water solutions enabled by fabricating heterogeneous thin films on three-dimensional substrates.

It has been extremely challenging to achieve the uniform distribution of a large volume fraction (>50%) of nanoscale fillers in a polymer matrix because the nanoscale fillers tend to agglomerate, and this agglomeration almost always degrades the mechanical properties of polymer nanocomposites. In this presentation, we show that it is possible to overcome this issue by using the modified atomic layer deposition that deposits metal oxides inside the polymer matrix instead of depositing them on the polymer surface. We use a SU-8 negative photoresist, which has a large-sized free volume due to its cage-like structure of the repeating molecular unit, so both the precursors (trimethylaluminum and diethylzinc) and water molecules can diffuse into SU-8 matrix during the atomic layer deposition process. Compared to the conventional atomic layer deposition, we intentionally used the longer purging time of precursor and water molecules to allow them to diffuse into SU-8 matrix. By repeating the infiltration process several times, it is also possible to control the volume fraction of nanoscale metal oxides. Transmission electron microscopy revealed that the infiltration depth is dependent on the type of precursor, and we obtained the ~50 nm thick infiltration depth of Al₂O₃ and over 150 nm thick infiltration depth of ZnO under the similar deposition condition. The high-resolution composition mapping showed that Al₂O₃ formed the particulate-like structure, but ZnO seemed to form the unique organic-inorganic interpenetrating network structure. We obtained the volume fraction of oxides by using the quartz crystal microbalance measurement, and both cases showed the uniform distribution of nanoscale oxides with ~44 vol.% of Al₂O₃ and ~24 vol.% of ZnO. Therefore, our synthesis technique enabled us to achieve the ideal microstructure of polymer nanocomposite successfully.
To evaluate the mechanical performance of our SU-8 nanocomposites, we performed in-situ nanomechanical testing on polymer nanocomposite nanopillars, which were fabricated by the e-beam lithography followed by the modified atomic layer deposition. Mechanical data revealed that our polymer nanocomposites exhibit metal-like strength (~400 MPa) but still polymer-like Young’s modulus (~8 GPa). This unusual combination of strength and Young’s modulus results in the giant modulus of resilience (~24 MJ/m³), the orders of magnitude higher than the modulus of resilience of most engineering materials. Modulus of resilience, the measure of a material’s ability to store and release elastic strain energy, is critical for realizing advanced mechanical actuation technologies in micro-/nano-electro-mechanical systems and in robotics. Our hybrid polymer nanocomposite features lightweight, ultrahigh tunable modulus of resilience, and versatile nanoscale lithographic patternability with potential for application as nanomechanical components, which require ultrahigh mechanical resilience and strength such as flexible display or durable three-dimensional structures in SU-8-based micro-electro-mechanical-systems (MEMS).

Results will be presented of our most recent studies of growth of Carbon Nanotubes (CNTs) by Chemical Vapor Deposition (CVD) from catalyst metals combined with heavy refractory diffusion inhibitor metals. The carbon source gas used in these studies is ethylene, C₂H₄, and various combinations of catalysts and diffusion inhibitor metals are investigated. The key concept is the inhibition of high-temperature erosion and Ostwald ripening of the catalyst particles by inclusion of the diffusion inhibitors, allowing the catalyst particles to have a longer lifetime and thus grow longer CNTs. We have previously demonstrated this effect using carbon monoxide (CO) as a carbon source gas, and we present here the extension of these studies to CNT growth using C₂H₄ as the source gas. Implications will be discussed for the growth of ultra-long (tens of cm or more) CNTs.

Sequential infiltration synthesis (SIS) is a modified atomic layer deposition process to precisely grow inorganic solids in soft material templates (e.g., polymers). Removal of the template via annealing results in the formation of polycrystalline inorganic thin films that inherit the morphology of the template matrix, offering opportunities to create a variety of nanostructured inorganic solid thin films. The properties of the resultant inorganic thin films rely heavily on the growth of the inorganic clusters in the template matrix. Therefore, to gain full control of the final inorganic thin films, it is of great importance to map out how the inorganic clusters grow inside the polymer templates.

In this work, structures of SIS-derived initial nucleating and final annealed indium-based inorganic materials with a PMMA template were systematically investigated as a function of SIS cycles using multiple tools, such as high-energy X-ray scattering (HEX) with pair distribution function (PDF) analysis, extended X-ray absorption fine structure, and transmission electron microscopy (TEM). The results resolved structures for the initial nucleating InO_xH_y clusters in PMMA as a function of the number of SIS cycles, which were found to exhibit similar local structures of cubic In₂O₃ and In(OH)₃ crystals. PDF analysis demonstrated that the SIS-derived InO_xH_y clusters in the first SIS cycle are characterized by a combination of mono-, di-, and tri-nuclear structures which and grow to larger clusters with higher aspect ratios during subsequent SIS cycles. Further extended SIS cycles lead to the increased number density of InO_xH_y clusters, which then condense to form larger and three-dimensional networks in the PMMA matrix. Annealing in air removes the PMMA template and converts the amorphous InO_xH_y clusters into crystalline In₂O₃ particles, which also show SIS-cycle-dependent properties.

Overall, this work contributes to the fundamental understanding of how inorganic clusters grow inside polymer matrices during the early stages of the SIS process.

High refractive index polymers (HRIPs) have recently emerged as an important class of materials for use in a variety of optoelectronic devices including image sensors, lithography, and light-emitting diodes. However, achieving polymers having refractive index exceeding 1.8 while maintaining full transparency in the visible range still remains formidably challenging. Here, we present a unique one-step vapor-phase process, termed sulfur chemical vapor deposition, to generate highly stable, ultrahigh refractive index (n > 1.9) polymers directly from elemental sulfur. The deposition process involved vapor-phase radical polymerization between elemental sulfur and vinyl monomers to provide polymer films with controlled thickness and sulfur content, along with the refractive index higher than 1.95. Notably, the HRIP thin film showed unprecedented optical transparency throughout the visible range, attributed to the absence of long polysulfide segments within the polymer, which will serve as a key component in a wide range of optical devices.

Despite the extensive research and available commercial memory products based on the phase-change-material (PCM), the Ge-Sb-Te alloy-based PCM’s full functionality for low-power and high-density memory has not been accomplished. This is mainly due to the lack of conformal film growth techniques, such as atomic layer deposition (ALD), of the chalcogenide-based PCM. The current PCM architecture relies on the crossbar arrays (CBA) structure due to its condensed and straightforward architecture. However, the CBA cannot keep up with the ultra-high density demands due to the increased fabrication cost of lithography steps and masks required for each stacking layer. It can be best addressed by the state-of-the-art vertical NAND architecture, proven technology with a stack of more than 150 layers. To fabricate highly integrated vertical-type crossbar array (V-CBA) memory, the Ge-Sb-Te phase change layer should be uniformly deposited in the inner surface of the deep etched holes vertical architectures, which could be accomplished only by ALD. Meanwhile, the high power consumption of the Ge-Sb-Te PCM during the write operation also poses another severe drawback.

This work demonstrates the development of GeTe/Sb₂Te₃ superlattice film utilizing ALD to achieve a high-performance V-CBA memory device. The chemical affinity of the ALD precursor to the substrate surface and the two-dimensional nature of the Sb₂Te₃ enabled to achieve the in-situ crystallized superlattice film where the out-of-plane directions of the two-component layers precisely matched the c-axis, while the in-plane directions were random. Such a c-axis-oriented Sb₂Te₃ could serve as the seed for the in-situ crystallized 111-direction aligned GeTe layer, which would have been either amorphous or randomly oriented polycrystalline layer if it was directly grown on a SiO₂ or TiN surfaces. The peculiar alignment method can be applied to deposit other van der Waals films such as transition metal chalcogenides and topological insulators with such alignments for future electronic device applications. The well-aligned and crystallized GeTe/Sb₂Te₃ superlattice film showed a RESET current of ~1/7 of randomly oriented Ge₂Sb₂Te₅ alloy films grown by the identical ALD process. The fluent improvement in the resistance switching performance from this GeTe/Sb₂Te₃ superlattice originates from the pressure-induced solid-state amorphization by the well-oriented superlattice films. The reduction of RESET current with the aligned GeTe/Sb₂Te₃ superlattice film was also feasible in the vertical structures, showing a step forward improvement for fabricating the futuristic V-CBA memory with low operating power and high density.

Hexagonal boron nitride (h-BN) is of great interest and potential due to its excellent electrical, optical, and mechanical properties and is isostructural to graphene. hBN is also a wide bandgap semiconductor with an optical bandgap of ~ 6 eV. To be used in the semiconductor industry, wide-area processing and engineering the bandgap is of utmost importance. As of now, chemical vapor deposition (CVD) is the most promising method to synthesize h-BN as it provides great control over the thickness of the deposited films and is the best candidate for the industrial scale-up process. To engineer the bandgap, in-situ doping is the most convenient and the most effective way, as it offers great control over the dopant material in terms of selection, introduction into the chamber, controlling the dopant concentration. Synthesizing hBN with in-situ doping directly on silicon, which is the most commonly used substrate in the semiconductor industry, eliminates the need for a post-synthesis transfer process and the employment of another process for doping. In this presentation, we have shown the direct synthesis of Mg-doped hBN films on silicon by chemical vapor deposition. XPS analysis confirmed the incorporation of Mg in hBN films. UV-vis spectroscopy was utilized to find the bandgap of the doped hBN films. As expected, it was concluded that the bandgap of the Mg-doped hBN films was less than the bandgap of the undoped hBN films.

Despite its importance for many industries, a quantitative model for the volatility of inorganic compounds has remained elusive. Vapor-based deposition technologies, including CVD and ALD, make use of certain metalorganic complexes and related compounds that show appreciable vapor pressure under moderate heating. However the rational discovery of new complexes is hindered by a lack of detailed understanding to link chemical identity with the conditions for evaporation or sublimation. To overcome these issues, we present a model for volatility and discuss its accuracy and scope.

We have assembled experimental pressure-temperature data from the literature for over 1,000 inorganic complexes, featuring 70 metal centers and 400 different ligands. We use machine learning (ML) approaches to train surrogate models for the dependence of evaporation, boiling or sublimation temperature on chemical identity and pressure. A variety of ML algorithms (Random Forest, Neural Networks etc) are studied in conjunction with different chemoinformatic descriptors and fingerprints. When supplied with the chemical identity and topology of a candidate inorganic compound, the trained ML model is capable of predicting the evaporation temperature at a specified pressure with a mean error of 10 degrees C for the most common metals. The speed with which the model is evaluated makes this approach suitable for screening thousands of candidate complexes, as one step within computational and/or experimental discovery of new precursors.

Polymer thin films can be formed conformally on a substrate by chemical vapor deposition (CVD). Specifically, initiated and oxidative chemical vapor deposition (iCVD and oCVD) can easily form nanometer-thin polymer films conformally on a wide range of substrate topologies and materials, from planar silicon, glass or metal to microporous membranes, fabric and paper to nanoporous particle networks, tubes and fibers. In fact, one of the major advantages of iCVD and oCVD is that they are able to use mild reaction conditions, including mild substrate temperatures (at or near room temperature) and mild vacuum (using only a roughing pump), to conformally deposit on such a varied range of substrate geometries and materials. As such, the solid substrate itself on which the polymer film is deposited has generally not been considered an important or determinant factor in influencing the quality and characteristics of the deposited film.

Here, in this talk, the role of the substrate in influencing the deposition behavior and characteristics of the polymer films, particularly for depositing nanometer-thin films by iCVD and oCVD, will be scrutinized more carefully. Specifically, we will present evidence that indicate the surface energy of the substrate can have a profound impact on the resulting conformality of the deposited polymer film. In addition, we will present results that further indicate the substrate temperature can yield different rate-limiting regimes that impact polymer growth rates, where previously it has generally been considered that deposition rate is only limited by monomer adsorption or availability of the monomer reactant at the substrate surface. These new insights challenge our current understanding and notion of iCVD and oCVD polymerization, and can offer new avenues and opportunities for iCVD and oCVD processing and their applications.

In recent years, considerable efforts have been done in the development of new functional materials for the fabrication of increasingly sophisticated and advanced photovoltaic devices (PV) that can ensure an efficient exploitation of solar spectrum. An advanced approach used to enhance the efficiency of PV systems, is based on the conversion of photons with energies outside the absorption range of the photoactive material (usually silicon) into a more suitable optical region. In this context, lanthanide (Ln)-doped fluoride materials are considered particularly promising for energy conversion (EC) applications due to their unique luminescent properties under light irradiation. The EC process can be achieved by simply integrating, into the PV layers, a material able to host an active luminescent Ln element. Upconversion (UC) and downconversion (DC) are the main luminescence mechanisms used for EC applications which increase the efficiency of photovoltaic devices. Concerning different Ln elements used for DC processes, europium has attracted great attention from the scientific community, thanks to its high peculiar chemical properties, especially in the oxidation state +3. Up to now, for EC processes it has been demonstrated that fluorides, compared to other inorganic matrices are more efficient inorganic hosts. In particular, BaF₂ is considered a suitable host for the UC and DC luminescence of Ln³⁺ ions because of its low phonon energy (similar or even lower than the extensively used hosts such as NaYF₄), high chemical stability, wide transparency (i.e. BaF₂ is transparent from 150 nm to 15 μm) and versatile synthetic strategies. It is important to stress that host materials require lattice matches with Ln³⁺ dopant ions. Trivalent lanthanide ions can be well incorporated into crystal structure of alkaline-earth fluorides due to the similar coordination sphere. However, a charge compensation mechanism is required to balance the additional charge introduced after the replacement of divalent alkaline-earth ions with trivalent Ln dopants.
In the present study, we report the synthesis of Eu³⁺-doped BaF₂ thin films on silicon and quartz substrates, using metalorganic chemical vapor deposition technique (MOCVD) focusing the attention on the fabrication process and the DC luminescence properties. The advantages of the MOCVD approach includes simplified instrumental apparatus, high deposition rate, high purity and easy industrial scale-up. For the fabrication of BaF₂ thin films doped with Eu³⁺, a multicomponent mixture of metalorganic adducts in an appropriate molar ratio is used as a single molten source. It is possible to easily tune the chemical compositions of the films, to tailor the energy transfer processes, by changing the composition of the precursor mixture, that is the Ba:Eu ratio in the multicomponent source. X-ray diffraction analysis (XRD), field-emission scanning electron microscopy (FE-SEM) and energy dispersive X-ray analysis (EDX) allowed a structural, morphological and compositional characterization of the films. Finally, the film functional properties have been assessed through luminescence spectroscopy.

The scientific research activity on AlGaN/GaN-based high-electron-mobility-transistors (HEMTs) is always under continuous and progressive development because of their promising performances for future electronic devices ,[1,2]. However, the conventional HEMTs with Schottky contact between the metal gate and AlGaN exhibits some reliability issues related to leakage current and current collapse phenomena, but also to limited gate voltage swings and possible metal inter-diffusion phenomena during the high temperature device processing. In this context, the realization of Metal-Insulator-Semiconductor (MIS)-HEMTs structure, with high-κ insulator integration between the metal gate and AlGaN, is the most favorable method to solve these issues [3]. Among the different kinds of high-k dielectrics (Al₂O₃, HfO₂, ZrO₂, AlN), the aluminum oxide (Al₂O₃) is the most widely reported for the GaN focused technology. The large Al₂O₃ implementation is closely related to the excellent physical properties as insulator, namely high permittivity (κ ~ 9), high critical electric field (10 MV/cm), large band gap (~8.9) and advantageous conduction band alignment with AlGaN/GaN structure. The ALD is the growth technique of choice for microelectronics high-k gate insulators, and in particular, in the case of Al₂O₃ thin film growth on AlGaN/GaN heterostructures a debate aspect is related to the ALD growth modes to be implemented [4,5]. In fact, some works reported on quite promising physiscal properties by thermal ALD process, while many others claimed the best choice is the plasma enhanced-ALD approach. In this context, we compared of the ALD growth modes of Al₂O₃ on AlGaN/GaN by both thermal (ALD) and plasma (PE-ALD) approaches. In particular, the deposition very thin Al₂O₃ layers have been investigated since the early growth stages (about ~1 nm thick layers) and upon increasing thickness up to 6 nm. Investigation has been carried out by both X-ray photoelectron spectroscopy (XPS) and Conductive- Atomic Force Microspopy (C-AFM), in order to correlate the physical covering upon film growth to the insulating properties of the growing Al₂O₃ layers. Two different growth mechanisms have been found, i. e. a quite ideal “layer-by-layer” growth in the case of PE-ALD process, while the thermal ALD process is quite consistent with an “island growth modes”. C-AFM can also provided nano-scale electrical information on the intimate electrical properties of the ultra-thin dielectrics and besides the standard imaging, it has been used for statistical study on Al₂O₃ thin films reliability upon stressing applied voltages on nanoscale matrix of punctual I-V measurements. The best performance has been entirely obtained for PE-ALD grown Al₂O₃ films both for the complete covering of the substrate surface and for the total and homogeneous insulating properties.

[1] U. K. Mishra, P. Parikh, and Y. F. Wu, Proc. IEEE 90, 1022 (2002)
[2] K. J. Chen and C. Zhou, Phys. Sta. Sol. (a), vol. 208, no. 2, 434–438, (2011)
[3] F. Roccaforte, P. Fiorenza, G. Greco, R. Lo Nigro, F. Giannazzo, F. Iucolano, M. Saggio, Microelectronic Engineering 187–188, 66–77, (2018)
[4] E. Schilirò, R. Lo Nigro, P. Fiorenza, F. Roccaforte, AIP ADVANCES6, 075021 (2016)
[5] E. Schilirò, P. Fiorenza, C. Bongiorno, C. Spinella, S. Di Franco, G. Greco, R. Lo Nigro, F. Roccaforte, AIP Advances 10, 125017 (2020)

Ideal properties for transparent thin film heaters (TTFHs) can be summarized as low sheet resistance, high transmittance in visible range and high thermal stability. Generally, the material of choice converts the electrical energy into heat by “Joule Heating Mechanism”. Therefore, achieved maximum temperature is related with applied voltage, sheet resistance and heat loss. Also, uniformity of such material is important due to heating consistency and stability of the entire film. Oxide films grown by atomic layer deposition (ALD) has advantages such as high uniformity, thickness control at angstrom level and excellent step coverage due to its self-limiting growth mechanisms¹. Aluminum doped zinc oxide (AZO) thin films growth by ALD expected to show ideal properties for a TTFH. The studies suggest that the AZO thin films are suitable to be used as heater on different substrates with different production methods, reason being chemical and thermal stability of AZO^2,3.
In this work, Al doped and undoped ZnO thin films grown by ALD were demonstrated as TTFH. The growth of the thin films were done on quartz substrates. Growth temperature for AZO was varied between 175 and 225 °C, Al doping was kept between 2-4 atomic percent, and thickness was kept constant as 100 nm for all samples. The varying temperature and doping levels were utilized to establish relation between doping concentration of aluminum on the performance of the thin films. The variables for ZnO were pulse/purge variations, and growth temperatures. Growth temperature for ZnO was in between 125 and 150 °C, and DI water pulse and Diethylzinc (DEZ- Zn(C₂H₅)₂) pulse was selected as 15 ms and purge time was kept at 10s. Different growth temperatures were chosen to vary stoichiometry of ZnO. Different temperatures were utilized to establish relation between ZnO thin films. Also, one of the main purposes was comparing Al dopped and undoped ZnO thin films with similar sheet resistance to display Al effect to TTFH performance such as maximum temperature and overall responsiveness. The lowest resistivity values achieved with AZO sample with 3.5 Al atomic percent as 3.3 10^-3 Ω.cm while the each produced thin films transparency were over 85% in spectral range. The expected maximum temperature values were over 70°C with applied fairly low voltage (≤10V) with heating rate of 20 °C/s.

1. George SM. Atomic layer deposition: An overview. Chem Rev. 2010;110(1):111-131. doi:10.1021/cr900056b
2. Ke S, Xie J, Chen C, et al. Van der Waals epitaxy of Al-doped ZnO film on mica as a flexible transparent heater with ultrafast thermal response. Appl Phys Lett. 2018;112(3). doi:10.1063/1.5010358
3. Tonbul B, Can HA, Ozturk T, Akyildiz H. Solution processed aluminum-doped ZnO thin films for transparent heater applications. Mater Sci Semicond Process. 2021;127(February). doi:10.1016/j.mssp.2021.105735

In recent years, copper-oxide compounds represent one of the most studied classes of semiconducting materials due to their main advantages of easily tuning the optical and electronic properties within the semiconducting behavior. For this reason, copper oxide materials play a significant role in many technological applications, ranging from batteries to catalysis, from photodetectors to solar cell devices. In particular, among the three most common and stable copper oxide phases, Cu₂O, Cu₄O₃, and CuO, named cuprite, paramelaconite and tenorite, respectively, the cuprite phase (Cu₂O) has been intensively studied and applied as p-type semiconducting material in many technological fields. In fact, Cu₂O represents an attractive material, due to copper abundance, nontoxicity, a wide variety of synthetic strategies and low-cost production.
Herein, we report an in-depth study of the metalorganic chemical vapor deposition (MOCVD) process for the fabrication of both cuprite Cu₂O and tenorite CuO copper oxide thin films starting from β-diketonate copper (II) precursor, i.e. Cu(tmhd)₂ bis(2,2,6,6-tetramethyl-3,5-heptanedionate). Operative conditions of fabrication strongly affect both composition and morphologies of the copper oxide thin films. Deposition temperature has been accurately monitored in order to stabilize and to produce selectively and reproducibly the two phases of cuprite Cu₂O and/or tenorite CuO. The present approach represents a simple, easily scalable and industrially appealing process for the production of compact and homogeneous copper oxide films at relatively low temperature. X-ray diffraction (XRD), field-emission scanning electron microscopy (FE-SEM) and atomic force microscopy (AFM) analyses allowed an accurate determination of the physico-chemical properties of deposited layers.
Moreover, as preliminary studies, two main applications have been tested: i) the use of Cu₂O thin films as hole transporting layer for the hybrid perovskite (methylammonium lead iodide, MAPI) based solar cell devices; ii) a core-shell system of all-oxide ZnO-Cu₂O n-p junction as fast and stable self-powered photodetectors. Morphology, crystalline and electronic structure of the Cu₂O thin films materials play a crucial role in determining the main functional parameters of both applications.

Infiltration of polymer templates with inorganic precursors using ALD technique enables the design of composite and all-inorganic structures for a broad range of applications. We will present the strategies for fabrication of conformal coatings with tunable thicknesses and porosities that reveal promising antireflective properties,[1] thermally stable multifunctional catalysts[2] and highly sensitive sensing layers.[3] We will present a systematic study on the conditioning of the polymer templates for its effective use for targeted applications. We will show that combining the infiltration of the polymer template from solution and from gas phase can significantly expand the compositional diversity of the structures that can be generated via ALD processes. We will discuss the perspectives of using plasma and thermal ALD for polymer infiltration. We will show that the choice for postpreparative polymer removal techniques can affect the dimensions, structure and composition of deposited all-inorganic heterostructures.[4]

[1] D. Berman, S. Guha, B. Lee, J.W. Elam, S.B. Darling, E.V. Shevchenko, Sequential Infiltration Synthesis for the Design of Low Refractive Index Surface Coatings with Controllable Thickness, ACS Nano 11 (2017) 2521-2530.
[2] Y. She, E.D. Goodman, J. Lee, B.T. Diroll, M. Cargnello, E.V. Shevchenko, D. Berman, Block-Co-polymer-Assisted Synthesis of All Inorganic Highly Porous Heterostructures with Highly Accessible Thermally Stable Functional Centers, ACS Applied Materials & Interfaces 11 (2019) 30154-30162.
[3] D. Berman, E. Shevchenko, Design of functional composite and all-inorganic nanostructured materials via infiltration of polymer templates with inorganic precursors, Journal of Materials Chemistry C (2020).
[4] D. Berman, Y. Sha, E.V. Shevchenko, Effect of Polymer Removal on the Morphology and Phase of the Nanoparticles in All-Inorganic Heterostructures Synthesized via Two-Step Polymer Infiltration, Molecules 26 (2021) 679.

Multiferroics are materials in which at least two of the three ferroic orders, ferroelectricity, ferromagnetism or antiferromagnetism and ferroelasticity coexist. Among multiferroics, perovskite bismuth ferrite (BiFeO₃) and its derived systems are of special interest for keeping their properties in extreme temperature environments because of their Curie (T_C = 1103 K) and Neel temperatures (T_N=643 K) well above room temperature. A lot of research has been devoted to the optimization of BiFeO₃ as ceramics, thin films and nanostructures. The interest in BFO is also boosted by it being a lead-free perovskite since the rising importance of environmental issues and questions of process sustainability have contributed to the prominence of lead-free perovskite systems. Photoelectric, pyroelectric, and piezoelectric properties are the most studied and appealing characteristics, which make BiFeO₃ an important material for energy harvesting applications. The possibility of combining the above-mentioned properties in a single device, a hybrid energy harvester, makes BiFeO₃ one of the most promising materials for the next generation of lead-free harvesters. The main deposition techniques applied to the production of BFO films are chemical solution deposition (CSD), pulsed laser deposition (PLD), sputtering and sol-gel. So far, compared to other methods, Metal-Organic Chemical Vapor Deposition (MOCVD) has been relatively unexplored and never used to develop BiFeO₃ deposition on silicon to fit the standard of the MEMs industry. MOCVD deposition technique offers a large number of advantages compared to other thin-film production methods with a wide choice of industrial precursors, high-quality substrates, and upscale opportunities.
In this presentation, a simple synthetic strategy based on MOCVD for the fabrication of pure and doped BiFeO₃ perovskite thin films on Si/IrO₂ and STO:Nb (100) substrates will be presented. The carried work aims to bring a methodology for the synthesis of BiFeO₃ systems thin films that are compatible with conventional industrial processes. The MOCVD approach proved a powerful tool for the fabrication of Dy-doped BiFeO₃ films. An accurate control of the processing parameters, in particular the multicomponent source mixture ratio, resulted selectively and reproducibly in the formation of Bi_1−xDy_xFeO₃ (x = 0, 0.06, 0.08 and 0.11) thin films. Ferroelectric, dielectric and pyroelectric properties of Bi_1−xDy_xFeO₃ (x = 0, 0.06, 0.08 and 0.11) thin films have been investigated showing that this films may be integrated as lead-free pyroelectric energy harvesters.
As material chemical composition has a critical influence on its properties, precursor thermal behavior has been checked by thermogravimetric analysis (TGA) for complex doped systems. The understanding of the films chemical compositions and homogeneity has been obtained by energy-dispersive X-ray analyses (EDX) and X-ray photoelectron spectroscopy (XPS). The polycrystalline or epitaxial nature of the films has been controlled by X-ray diffraction (XRD) and transmission electron microscopy (TEM) in the case of intricate nanostructures. Morphological study of the materials has been done by field emission scanning electron microscopy (FE-SEM) and atomic force microscopy (AFM). Coupled to the AFM, piezoresponse force microscopy (PFM) and piezoresponse force spectroscopy (PFS) have permitted the imaging of ferroelectric domains and the confirmation of the films piezoelectric properties.


This work was supported by the European Community under the Horizon 2020 Programme in the form of the MSCA-ITN-2016 ENHANCE project, Grant Agreement N.722496.

Oxides are widely appealing, owing to their high mobility (compared to organic materials) and robust electronic properties. Whilst less common than their n-type counterparts, p-type oxides are needed as hole transport layers in photovoltaics, as well as for CMOS. For next-generation applications, it is critical to deposit p-type oxides at low temperature in order to be compatible with thermally-sensitive underlayers, such as flexible substrates (e.g., for wearable electronics) and photovoltaic device stacks (e.g., based on lead-halide perovskites).

This talk explores the use of atmospheric pressure chemical vapor deposition (AP-CVD) for the growth of p-type cuprous oxide (Cu₂O) at low temperature for photovoltaic and field-effect transistor (FET) applications.

In the first part, the deposition of uniform Cu₂O by AP-CVD at temperatures down to 100 °C is developed. The growth rate is 1.3 nm s^-1, which is an order of magnitude larger than conventional atomic layer deposition. This is used to create the p-type buffer layers on n-i-p structured perovskite photovoltaics. We show that owing to the high density of the Cu₂O films grown, a 3 nm layer of Cu₂O is sufficient to mechanically protect the perovskite device stack from damage when sputter depositing a transparent indium tin oxide electrode on top. The Cu₂O layer has minimal parasitic optical absorption (with >95% transmittance) and hole mobilities of 4±2 cm² V^-1 s^-1, which is orders of magnitude larger than that of the PTAA hole transport layer in the perovskite device stack. This leads to semi-transparent perovskite photovoltaics with state-of-the-art efficiencies of 16.7%. By stacking these semi-transparent devices over silicon bottom cells made with industrially-relevant processes, four-terminal tandems with efficiencies up to 24.4% are achieved.^[1]

In the second part of the talk, AP-CVD Cu₂O is developed into FETs. The switching characteristics of the devices are examined and compared against Cu₂O grown by atomic layer deposition under comparable processing conditions.^[2] The effect of Al₂O₃ passivation on the performance of the FETs is investigated, coupled with in-depth photoemission spectroscopy and transmission electron microscopy to understand the cause of changes in performance.

References

[1] Jagt, …, Hoye, ACS Energy Lett., 2020, 5, 2456
[2] Napari, et al., ACS Appl. Mater. Interfaces, 2021, 13, 4156

Low temperature electrolysis of water as a source of protons and electrons for electrochemical synthesis of hydrogen, liquid fuels and chemicals is an essential approach for decarbonization of the global economy. Water electrolysis is heavily reliant on highly active and stable electrocatalysts, and the most challenging step, from the perspective of electrochemical kinetics and catalyst stability is the oxygen evolution reaction (OER). High-rate water oxidation leads to acidic conditions in the electrolyte near the anode surface in electrolyzers. Among the highly active OER catalysts, only iridium oxide is relatively stable at acidic pH. However, the efficient utilization of IrO_x-based catalysts is essential for a cost-effective and scalable implementation of water electrolysis systems. The IrO_x catalyst layers need to be mechanically robust and stable during water oxidation and remaining highly active over long time periods.

Atomic layer deposition (ALD) is well-suited for the deposition of ultra-thin catalysts and support materials, and for catalysts surface functionalization to promote stability and enhance catalytic performance. We have synthesized stable nanoscale metal oxide alloys in the IrO_x-TiO₂ system that catalyze water oxidation while achieving high electronic conductivity. Thin TiO₂-IrO_x alloy films were grown by a super-cycle ALD process consisting of alternative cycles of TiO₂ and IrO₂ ALD processes. ALD TiO₂ films were grown using tetrakis(dimethylamido)titanium (TDMAT) and water vapor. IrO₂ ALD was performed by using ethyl cyclopentadienyl cyclohexadiene iridium (I) ((EtCp)Ir(CHD)) and ozone as the oxidant. The alloy compositions were controlled by varying the ALD cycle ratio. X-ray photoelectron spectroscopy (XPS) determined that the Ir/(Ir+Ti) compositions were reproducible and varied in a range from 0.13 to 0.51. Hall-effect analysis exhibits an overall decrease in the resistivity of the TiO₂-IrO_x alloys with increasing the iridium content, and p-type conduction for ALD-IrO₂ films.

Another approach to stabilize the catalyst layers is the application of overlayer coating for corrosion protection. The coating can be an electrochemically active overlayer or a combination of overlayer and catalyst material, which induces activity for catalyst reactions on the surface. It may also be a nominally inactive material, which allows the correct reaction species to come to the catalyst/electrolyte interface, while preventing other components poisoning the surface of the catalyst. We observed coating of ALD-IrO_x catalyst layers with very thin layers (< 1 nm) of TiO₂ which is normally an inactive material for water oxidation, enhances its activity by reducing the activation overpotential for OER. The current associated with water oxidation (electron transfer in OER) at a given potential was higher when we had a thin TiO₂ coating with a thickness on the order of less than one nanometer. We report on possible chemical modification of the IrO_x surface caused by the TiO₂ coating and how this correlates with the observed overpotential and Tafel behavior for OER.

Boron nitride nanotubes (BNNTs) are a new type of one-dimensional material that is similar to carbon nanotubes. BNNTs have recently attracted attention due to their desirable physical, thermal, and mechanical properties. Due to the high surface area and stability under harsh conditions, BNNTs can be a suitable substrate supporting catalyst particles, including for use in harsh, plasma-assisted reactors. In this research, Ni nanoparticles are deposited on BNNT papers via plasma enhanced atomic layer deposition (PEALD) using bis ethylcyclopentadienyl nickel (Ni(EtCp)₂) as the precursor and N₂/H₂ plasma. Scanning Electron Microscopy (SEM) and Helium Ion Microscopy (HIM) studies show BNNTs to be well covered with Ni particles after 150 cycles of ALD at 350 ^oC. Elemental analysis (EDS) proves the existence of Ni on the BNNT papers after the ALD process. According to the Transmission Electron Microscopy (TEM) observations, Ni nanoparticles are uniformly distributed over the BNNTs and have a diameter of about 4 nm. The synergy arising from the catalytic nature of the Ni nanoparticles and the high surface area, porosity, and stability of the BNNT substrates makes this material system a perfect candidate for catalysis applications such as for the plasma-assisted production of ammonia.

Solid state batteries (SSBs) are considered the most promising next-generation energy storage technology both due to their increased safety over conventional wet Li-ion cells and their ability to fabricate novel architectures with simultaneous high energy and power density. However, the ionic conductivities of many solid electrolyte materials are orders of magnitude lower than liquid carbonate-based electrolytes. In order to offer SSBs with competitive charge/discharge rates, SSB layers need to be thin (for fast ion diffusion) but have a large surface area (to enable large ion flux from anode to cathode). These geometric constraints establish the need for 3D electrode (and solid electrolyte) structures for a high performance battery architecture. Atomic layer deposition (ALD) is ideally suited to fabricate such devices due to its low deposition temperature, highly conformal nature, and Angstrom-scale thickness control, however many challenges remain with materials fabrication and process integration.

We have developed both plasma-enhanced ALD (PEALD) and thermal ALD processes for the Li, Na, and K alkali polyphosphazene (A*PON) electrolytes^1,2,3,4. The PEALD A*PON processes, by virtue of their 4-precursor synthesis route, allow some degree of chemical composition tuning. The thermal A*PON processes, unlike their PEALD counterparts, have a predetermined P-N ratio fixed by use of the precursors, however there is a tradeoff between the low temperature deposition enabled using plasma-based A*PON processes, and the conformality offered using the thermal A*PON processes. I will discuss and compare the reaction mechanisms and growth behavior among the entire family of A*PON solid electrolyte thin films, as well as discuss the integration challenges on the path to developing all-ALD planar and 3D SSBs.

1. Pearse, A.; Schmitt, T.; Sahadeo, E.; Stewart, D. M.; Kozen, A.; Gerasopoulos, K.; Talin, A. A.; Lee, S. B.; Rubloff, G. W.; Gregorczyk, K. E., Three-Dimensional Solid-State Lithium-Ion Batteries Fabricated by Conformal Vapor-Phase Chemistry. ACS Nano 2018, 12 (5), 4286-4294.
2. Pearse, A. J.; Schmitt, T. E.; Fuller, E. J.; El-Gabaly, F.; Lin, C.-F.; Gerasopoulos, K.; Kozen, A. C.; Talin, A. A.; Rubloff, G.; Gregorczyk, K. E., Nanoscale Solid State Batteries Enabled by Thermal Atomic Layer Deposition of a Lithium Polyphosphazene Solid State Electrolyte. Chemistry of Materials 2017, 29 (8), 3740-3753.
3. Nuwayhid, R. B.; Jarry, A.; Rubloff, G. W.; Gregorczyk, K. E., Atomic Layer Deposition of Sodium Phosphorus Oxynitride: A Conformal Solid-State Sodium-Ion Conductor. ACS Applied Materials & Interfaces 2020, 12 (19), 21641-21650.
4. Kozen, A. C.; Pearse, A. J.; Lin, C-F.; Noked, M.; Rubloff, G. W.; Atomic Layer Deposition of the Solid Electrolyte LiPON. Chemistry of Materials 2015, 27 (15), 5324-5331.

Theoretically, the integration of nanomaterials, such as carbon nanotubes (CNT), into a continuous matrix will produce a nanocomposite material with new and exciting properties. However, problems with the dispersion of the nanomaterials in the matrix have greatly hindered the realization of this opportunity. For example, mixing a bulk sample of CNTs with a polymer matrix often requires shear forces high enough to damage the CNTs and the polymer in order to be able to get reasonable dispersion. Mixing a bulk sample of CNTs with a ceramic matrix, in order to make a ceramic nanocomposite, would be even more difficult. Fortunately, unlike polymers, ceramics can easily survive at the temperatures needed for chemical vapour deposition (CVD) of CNTs (and other nanomaterials). In this work, CNTs are deposited onto ceramic (alumina) powder by CVD after staining with catalyst precursor and jet milling. The CNTs are well separated from each other and immobilized on the powder. This hybrid CNT-ceramic powder is mixed with a binding agent, pressed into a green body, and pressurelessly sintered to fabricate the nanocomposite. Physicochemical characterization reveals the CNTs to have survived the sintering process. The nanocomposite is characterized for its mechanical properties and the process optimized. A “floating-catalyst”, plasma CVD reactor is under construction that will produce 1 kg/h of sinterable hybrid ceramic powder from a jet of neat alumina powder and a ferrocene / xylene solution, eliminating the need to deposit catalyst precursor and jet mill the powder.

Recent advances in technology accomplished by operating at nanoscale, require surface modification and functionalization for improved performance. However, functionalization of patterned surfaces or surfaces that are not directly exposed, present significant challenges in homogeniety and conformality. Functionalization of surfaces with highly conformal and homogenious coatings can be achieved via vapor deposition methods. Initiated chemical vapor deposition (iCVD) as a polymer thin film deposition technique provides conformal coatings of functional polymers on patterned surfaces.

In this talk, methods to obtain conformal iCVD coatings in confined spaces by optimizing the process parameters will be introduced and studies on one dimensional nanostructures and hybrid membranes prepared by templated iCVD method will be presented. The templated iCVD method involves conformally coating the pore walls of a sacrificial porous template with functional polymers. Stimuli responsive polymer nanotubes as nanocarriers for controlled release and biosensor applications, are successfully fabricated by templated iCVD method.

In the second part of this talk, smart membranes, which have tunable pore sizes that depend on the pH or temperature will be presented. Smart membranes are fabricated using the templated iCVD method and the conformality of the coatings are controlled by tuning the process parameters. The performance of these hybrid membranes are optimized for macromolecular separation and for protein gating applications.

With the growing interest in nanoscale materials for applications such as electronics, catalysts, and batteries, methods for fabricating materials with precise control are becoming increasingly important. Area selective atomic layer deposition (AS-ALD) is gaining attention as an important process to achieve spatial control of materials deposition down to the nanometer length scale. AS-ALD is a bottom-up patterning process in which material is deposited only where desired. This talk will describe how control over the chemical properties of the substrate surface can be used to achieve AS-ALD. A deep understanding of the gas-surface chemistry behind ALD can be critical toward developing area selective processes. Research into the fundamental chemical mechanisms of ALD will be illustrated, including studies into both ALD precursor design and surface treatments that can enhance or inhibit nucleation. The development of inhibitory or enhancing layers to alter the native surface reactivity will be described. We will show that this process provides good selectivity in the deposition of thin films ranging from inorganic to organic on a variety of substrate materials, including dielectrics and metals.

Storage of electricity into molecules is the most viable answer to the intermittency of renewable sources. In this respect, H₂O splitting to H₂ and O₂ is key to enable the usage of H₂ as fuel for transport, as well as building block together with CO for the electrosynthesis of hydrocarbons. For the O₂ evolution half reaction (OER), cobalt phosphate-based materials (CoPi) are interesting because they are based on earth-abundant elements and behave as bulk electro-catalysts, with Co³⁺ being the electro-chemically active centers.
Atomic Layer Deposition (ALD) is gaining major attention in the field of (electro-)catalysis because of its merit of control over conformality, uniformity, film growth (from nano-particles to thin films) and stoichiometry. Recently, we have contributed to this exciting field by demonstrating the synthesis of amorphous CoPi films by ALD [1]. CoPi is prepared by combining ALD of CoO_x from cobaltocene (CoCp₂) and O₂ plasma, with cycles of (CH₃O)₃PO (TMPO) and O₂ plasma, according to an ABCD recipe scheme [1]. The process shows a GPC of 1.12 ± 0.05 Å at 300^○C, linear growth and no nucleation delay. Insights into the surface reactions occurring during the ALD process are inferred by mass spectrometry [1]. The CoCp₂ dosing is characterized by the absence of by-products, suggesting that the adsorption of the precursor on the surface is likely to occur without ligand exchange. The subsequent O₂ plasma exposure promotes the combustion of the Cp ligands, as supported by the detection of CO₂⁺, CO⁺ and H₂O⁺ species. The presence of C₄H₄⁺ and C₂H₂⁺ species in the initial stages of the plasma step, provides indirect evidence that cyclopentadienone (C₅H₄O) is an intermediate product in the Cp oxidation process. Then, C₅H₄O dissociates, generating C₄H₄, C₂H₂ and CO. The TMP dosing is characterized by chemisorption accompanied by abstraction of one methanol molecule. Finally, the conversion of the remaining ligands occurs through the O₂ plasma exposure.
Next to the above-mentioned CoPi growth studies, we have also shown that tuning the Co-to-P ratio, by combining this recipe with extra cycles of CoO_x, leads to the increase in OER performance [2], beyond that achieved by traditional electro-deposited films.
Moreover, we have investigated the mechanism behind the enhancement of the catalytic activity of ALD CoPi when tuning the Co-to-P ratio. We have found that ALD CoPi thin films progressively undergo activation under alkaline conditions with increasing number of cyclic voltammetry (CV) cycles. During this activation process induced by the infiltration of the electrolyte in the bulk of the film, the current density increases in parallel with the leaching of phosphorus out of the electro-catalyst and the shift of the oxidation state of Co from Co²⁺ to Co³⁺ [3]. CV analysis combined with Rutherford backscattering measurements indicate that after activation, as much as 22% of all Co atoms become accessible to the electrolyte and are activated. The electrochemical surface area (ECSA), i.e. the surface area accessible to promote H₂O splitting, increases up to a factor of 30. The increase in ECSA is strongly dependent on the composition of the pristine CoPi films: while the aforementioned increase holds for CoPi films with a Co-to-P ratio of 1.6, for films with a Co-to-P ratio of 1.9 the ECSA only increases by a factor 3.6. We find that for all investigated Co-to-P ratios, the electrochemical activity scales linearly with ECSA upon activation. Thus, the pristine composition indirectly affects the activity of the electro-catalyst by guiding its restructuring during activation.
It can be concluded that the digital control over the chemical composition of ALD CoPi enables to unravel the mechanism behind its electrochemical activation.

[1] V. Di Palma et al., J. Vac. Sci. Technol. 38,022416 (2020)
[2] V. Di Palma et al., Electrochem. Commun. 98, 73 (2019)
[3] R. Zhang et al., ACS Catal.11, 2774 (2021)

Atomic layer deposition (ALD) of correlated oxides is an emerging field. With limited data available in literature on thermodynamics of precursors to deposit rare-earth metals, significant experimental effort and instrumental optimization is essential for robust formation of crystalline phases of complex oxides such as rare-earth nickel oxides. ALD of such complex semiconductors could make them directly relevant to electronic technologies as well as enabling high quality ultra-thin films necessary for electric field control of various physical properties. In this work, we will discuss ALD processing of epitaxial NdNiO₃ thin films. By controlling various deposition parameters including precursor pulsing time, deposition temperature, we have successfully optimized the growth conditions to obtain stoichiometric NdNiO₃ thin films grown epitaxially on LaAlO₃. We will then discuss their excellent functional properties such as large electrically tunable opacity to realize electrochromic window coatings on glass and highly responsive hydrogen gas sensors for environmental monitoring. Finally, we will critically compare these functional properties to those of similar thickness epitaxial NdNiO₃ films grown by RF-sputtering.

Photoelectrochemical (PEC) energy conversion is a promising route for synthesis of storable chemical fuels from sunlight, thereby providing a route to overcome the current global reliance on fossil fuels. One of the major challenges in such artificial photosynthetic systems are the poor efficiency and material instability of semiconductor photoelectrodes under harsh PEC operating conditions. One strategy to overcome this limitation is to interface the semiconductor light absorber with conformal and ultra-thin catalytic layers, which promote the desired chemical reaction, while permitting efficient interfacial charge transport, maintaining chemical stability, and minimizing parasitic light absorption. In this regard, plasma-enhanced atomic layer deposition (PE-ALD) is a powerful tool for designing surface layers and interfaces with tailored functionality and precise thickness control.

Recently, it was demonstrated that PE-ALD can be used to fabricate conformal, biphasic Co₃O₄/Co(OH)₂ catalyst layers on semiconductor photoelectrodes, which are simultaneously robust and electrochemically active¹. In this system, the nanocrystalline Co₃O₄ layer forms a durable interface to the substrate and the disordered Co(OH)₂ surface layer significantly improves the electrocatalytic oxygen evolution reaction (OER) activity. Here, we leverage the precise control of PE-ALD to further tailor the thickness ratio of the surface and interface layers of the Co₃O₄/Co(OH)₂ bilayer by tuning the plasma pulse time during growth. Short pulses lead to the formation of porous, unstable Co(OH)₂ layers, while long pulses reduce the thickness of the surface layer and form stable, inactive Co₃O₄ layers. Furthermore, we show that the underlying mechanism for the formation of the Co(OH)₂ surface layer is the incomplete decomposition of the precursor at short pulse durations. This work highlights the use of PE-ALD as a promising approach for engineering catalyst/semiconductor interfaces to create efficient and stable photoelectrodes.

1. Yang, J. et al. A multifunctional biphasic water splitting catalyst tailored for integration with high-performance semiconductor photoanodes. Nat. Mater. 16, 335–341 (2017).

Germanium (Ge) is a promising candidate to serve as an infrared light emitter in the fully integrated, miniaturized biosensing systems where electronic and photonic components are realized on the same silicon (Si) chip in a CMOS-compatible fashion. Despite the inherently indirect band structure of Ge, its bandgap can be engineered to convert it into an efficient light emitter with the emission wavelength tuned into the mid-infrared range of the spectrum, which is suitable for biosensing applications. Among various techniques utilized to enhance the light emission efficiency of Ge, such as Sn alloying, n-type doping, and application of tensile strain, the latest one provides an additional advantage by enabling the development of the room-temperature-operated infrared light source with a predefined emission wavelength, where the emission wavelength is simply a function of the strain. In this work, we introduce a CMOS-compatible fabrication technique that allows obtaining single-crystalline tensilely-strained germanium microstructures, where both crystallization and tensile strain induction is achieved through a single annealing process. The fabrication procedure of the structures relies on the deposition of the germanium, silicon nitride, and silicon dioxide films all by using a physical vapor deposition, namely sputtering, technique, which is a room-temperature-operation, environmentally-friendly deposition technique as opposed to the chemical vapor deposition methods. The crystallization of the sputtered amorphous germanium is achieved via liquid phase epitaxy following a single rapid thermal annealing process, during which tensile stress is also introduced into the silicon nitride film capping the germanium. A subsequent photolithography process allows the strain transfer from the tensilely-stressed nitride film to the lithographically defined germanium microstructure. Strain, and therefore the emission wavelength, can simply be adjusted by varying the duration of the final wet etching process. The method is suitable for both the fabrication of uniaxially and biaxially tensile-strained germanium microstructures, where uniaxial and biaxial strain levels of around 3% and 1.5%, respectively, have been demonstrated and verified by Raman spectroscopy and micro-photoluminescence (micro-PL) measurements, as well as the finite element method (FEM) simulations. Overall, our results show that infrared light emitters based on tensilely-strained germanium microstructures suitable for the biochemical sensing applications can be realized on Si chips in a CMOS-compatible fashion utilizing cost-efficient and easy-to-use deposition (sputtering) and crystallization (liquid phase epitaxy) methods.

Synthesis of inorganic nanowires/nanotubes suspended in the gas through floating catalyst chemical vapour deposition (FCCVD) produces exceptional growth rates of 500 - 1000 micron/second, more than two orders of magnitude faster than conventional substrate processes. It leads to nanowire lengths > 100 microns and thus to the possibility of direct assembly into freestanding macroscopic networks as a continuous process. This work studies the different reaction paths controlling conversion and crystallinity in FCCVD applied to the synthesis of silicon nanowires (SiNW) from silane, grown through an aerosol of gold catalyst nanoparticles. There are two main competing reactions: catalysed growth of SiNW and nucleation of amorphous Si nanoparticles. The two populations can be conveniently distinguished for high-throughput screening by their colour and Raman spectra. Their balance is kinetically controlled by the ratio of precursor/hydrogen carrier gas, through its inhibition of the pyrolisis of silane into silylene. This role of hydrogen is common to other nanomaterials grown by FCCVD, but differs with respect to carbon and boron nitride in having little influence on the nucleation stage of 1D nanomaterial growth. As a consequence, we were able to identify effective process conditions for producing large free-standing and pristine SiNW sheets. These yield highly promising results when exploited as active material in lithium ion battery anodes.

Metal oxides are commonly used as sensing material for gas detection applications. The surface of the metal oxide is covered with oxygen ions (ion adsorbed) in presence of air, and these ions act as electron acceptors [1]. Metal oxides have mostly been used in resistive or conductive gas sensors, where electronic events in the material generate a detection signal. However, they are also promising as sensing layer in gravimetric micro-electromechanical systems (MEMS). MEMS-based gravimetric sensors are ultra-fast and highly sensitive and the sensing mechanism is related to mass change. Upon exposure of the metal oxide-coated sensor to a gas, the surface oxygen ions react with the gas molecules, which results in lowering or increasing the sensor mass [2]. The sensing performance of the sensor is highly dependent on the chemical and structural properties of the sensing layer. Recently, Atmospheric-Pressure Spatial Atomic Layer Deposition (AP-SALD) has proven to be an excellent deposition technique that is capable of producing high quality metal oxide thin films with precision control, while being up to 2 orders of magnitude faster than conventional ALD, and working at atmospheric pressure [3][4].
Recently, we developed ZnO-based transparent conductive films deposited with our home-made AP-SALD for resistive gas sensing [5]. In this research, we employed AP-SALD for deposition of different metal oxides for gravimetric gas sensing. We present a complete study on deposition of tin oxide (SnO₂), as a common sensing material, and structural, optical, chemical and electrical properties of the SnO_x thin films will be shown. Gas sensing properties of SnO₂, ZnO, Al-ZnO, etc. deposited by AP-SALD and effect of annealing and doping on sensing behavior will be investigated and discussed in detail. The gas sensors functionalized by metal oxides showed fast response and high sensitivity toward humidity, ethanol and acetone.
References:
1. M. E. Franke, T. J. Koplin, and U. Simon, Small 2, 36 (2006).
2. K. M. Goeders, J. S. Colton, and L. A. Bottomley, Chem. Rev. 108, 522 (2008).
3. D. Muñoz-Rojas and J. Macmanus-Driscoll, Mater. Horizons 1, 314 (2014).
4. K. P. Musselman, C. F. Uzoma, and M. S. Miller, Chem. Mater. 28, 8443 (2016).
5. V. H. Nguyen, D. Bellet, B. Masenelli, and D. Muñoz-Rojas, ACS Appl. Nano Mater. 1, 6922 (2018).

Cuprous oxide (Cu₂O) is a non-toxic and abundant p-type semiconductor with a direct band gap around 2.1 eV and a large visible absorption coefficient. It has been studied and developed for several devices such as solar cells, thin film transistors or batteries. In this work, Cu₂O think films with thickness below 100 nm and record motilities up to 92 cm2V^-1s^-1 have been obtained. The films have been deposited via AP-SALD (Atmospheric-Pressure Spatial Atomic Layer Deposition) at low temperatures (up to 260 °C), under atmospheric pressure. AP-SALD is an alternative approach to conventional ALD in which the precursors are separated in space rather than in time, allowing fast deposition rates as compared to conventional ALD (up to nm/s in some cases).
The impact of the deposition parameters on the growth rate and transport properties (mobility and concentration of carriers) has been studied, and different Cu precursors have been evaluated. The high mobility values obtained, which are close to values associated to epitaxial Cu₂O thin films or Cu₂O single crystals, have been rationalized based on Raman data. Optimized Cu₂O thin films, combined with n-type ZnO also deposited by AP-SALD, lead to all-oxide solar harvesters with efficiency values rivaling similar devices made with high temperature and/or vacuum approaches, thus opening the door to efficient semitransparent solar harvesters.

Cuprous oxide (Cu2O) is a promising material for emerging electronics. Bandgap and defect states estimation of Cu2O film grown by chemical vapor deposition (CVD) technique on SiO2/Si substrate at 600°C was performed in this work using various spectroscopies. XRD and XPS confirmed the phase purity of deposited film. The optical bandgap of Cu2O was 2.35 eV which was calculated from absorption spectroscopy (UV-Vis). Using ultraviolet photoelectron spectroscopy (UPS) and absorption spectroscopy (UV-Vis), the nature of deposited Cu2O was confirmed as a p-type semiconductor. Using UPS, the energy level of the valence band and Fermi level of the intrinsically p-doped Cu2O film were determined as 6.02 eV and 5.62 eV respectively from the vacuum level. The temperature dependence photoluminescence spectra of these films exhibited emission peaks at ~1.4 eV and ~1.72 eV which corresponds to energy states of copper and oxygen vacancies level in the bandgap. Photoluminescence spectra also show a peak at ~2 eV which is due to the relaxation of free excitons to the valance band.

Solid layers^1,2 of ZrO₂ are often deposited by chemical vapor deposition (CVD), where a metal-containing precursor is evaporated and forms a film in a surface reaction. To design such a process, the gas phase reactions, especially their temperature-dependent kinetics, should be known, which is seldom the case. Intermediates, formed in the gas phase can lead to film formation or to a depletion of the precursor and therefore a reduction in growth rate. The analysis of these early stages of growth requires fast and sensitive analytical techniques with sufficiently low detection limits for elusive gas phase species. Experimental limitations precluded the detection of most of the postulated intermediate species, and reaction mechanisms are predominantly hypothetical so far.^3,4 An experimental approach to enlighten the gas phase chemistry of an often used precursor, zirconium acetylacetonate, is presented here, using a microreactor coupled to a mass spectrometer with mild ionization from a synchrotron source. This was demonstrated previously, to be an effective way to detect and characterize gas phase intermediates with short lifetime.^5,6
A combined numerical and experimental study of the vacuum pyrolysis of zirconium acetylacetonate, Zr(C₅H₇O₂)₄, in a microreactor at short residence times of < 60 μs was carried out. The precursor is sublimed, subsequently transported by argon or helium carrier gas, and expanded through a pinhole into a resistively heated 1 mm inner diameter SiC-microreactor. Products leaving the reactor are expanded and ionized by tuneable vacuum ultraviolet (VUV) synchrotron radiation and characterized by imaging photoelectron photoion coincidence spectroscopy (i²PEPICO) and mass spectrometry at the Swiss Light Source in Switzerland. From the temperature-dependent photoionization mass spectra between 130 and 650 °C, two initial decomposition steps were identified and temperature regimes for the following pathways of thermal decomposition at higher reactor temperatures were determined. In addition, we recorded photoionization efficiency curves and threshold photoelectron spectra (TPES) at photon energies of 6.5-11.5 eV, which gave us direct evidence about the intermediates and products formed. Six important Zr-intermediates were identified for the first time, as for example Zr(C₅H₇O₂)₂(C₅H₆O₂), and ZrO(C₅H₇O₂)₂, together with several organic products, which aided us to describe the predominant thermolysis pathways in the temperature range of 130-650 °C. In conjunction with the simultaneous numerical simulation of the flow field in the microreactor, the application of synchrotron radiation coupled to the i²PEPICO experimental apparatus is a promising way to enlighten pyrolysis pathways and kinetics of metal-organic precursors.


References
1. A. Méndez-López, O. Zelaya-Ángel, M. Toledano-Ayala, I. Torres-Pacheco, J.F. Pérez-Robles and Y.J. Acosta-Silva, Crystals, 2020, 10(6), 454.
2. J.G. Thakare, C. Pandey, M.M. Mahapatra and R.S. Mulik, Met. Mater. Int., 2020, 366, 338.
3. M.V.F. Schlupp, J. Martynczuk, M. Prestat and L.J. Gauckler, Adv. Energy Mater., 2013, 3(3), 375.
4. H.M. Ismail, Powder Technology, 1995, 85(3), 253.
5. Z. Zhang, Y. Pan, J. Yang, Z. Jiang and H. Fang, Journal of Crystal Growth, 2017, 473(12), 6.
6. P. Hemberger, J.A. van Bokhoven, J. Pérez-Ramírez and A. Bodi, Catal. Sci. Technol., 2020, 10(7), 1975.