Symposium FF05—Advanced Atomic Layer Deposition and Chemical Vapor Deposition Techniques and Applications

Cobalt (Co) and Ruthenium (Ru) are used as a seed layer for metallization of interconnects, where atomic layer deposition (ALD) is applied to achieve conformality and precise thickness control at the atomic scale. Plasma-enhanced ALD (PE-ALD) is used for low-temperature thin film growth by alternating exposures of metal precursors and plasma reactants. During the PE-ALD growth of metals, N-plasma, for example, NH₃ or a mixture of N₂ and H₂, has been developed to avoid surface metal oxidation. In this presentation, we study the PE-ALD growth of Co and Ru by first principle calculations. Experimentally known surfaces were constructed including (001), (101), and (100). The (001) surface with a hexagonal structure is the most stable surface and the (100) surface with a zigzag structure is least stable but with high reactivity. The two surfaces were chosen to study the surface saturation coverage by considering individual adsorption and co-adsorption of NH and NH₂. On the (001) surface, the saturation coverage for NH is 1ML on Ru and 5/9ML on Co. For NH₂, the saturation coverages are 5/9ML on bpth Co and Ru surfaces. Additionally, NH₂ is unstable under high coverage by either desorbing from the metal surface or dissociating into NH. The calculated dissociation barrier for NH₂ dissociation on both Co and Ru surfaces is 0.71eV. On the (100) surface, the saturation coverages are 2ML for NH and 1 ML for NH₂ on both Ru and Co surfaces. The larger saturation coverage on (100) surfaces is attributed to the unique trench structure, which provides more available surface sites than that of (001) surfaces. We also consider co-adsorption of NH and NH₂ on (001) and (100) surfaces. The results are analyzed with ab initio thermodynamics by calculating the Gibbs energy. Both the ultra-high vacuum (UHV) condition and standard ALD operating condition are used to elucidate the effect of pressure and temperature on the termination of metal surfaces.
The adsorption and reaction of metal precursors (CoCp₂ and RuCp₂) on NH_x terminated metal surfaces was investigated with inclusion of van der Waals corrections.¹ Two possible adsorption structures, namely the precursor parallel and perpendicular to the NH_x-terminated surfaces, were considered. The adsorption and reaction of metal precursors are classified into four steps: adsorption, pre-reaction, proton transfer, and ligand desorption. The barrier for proton transfer is calculated using climbing image nudged elastic band (CI-NEB) method. After desorption of Cp ligand, the Co or Ru metal binds to the N atom that transferred the H atom to the Cp ligand. The remaining Cp ligand will be consumed in the second half cycle with the help of the N-plasma. After a full cycle, the surface is an NH_x-terminated metal surface and ready for the next cycle. In addition, the NH_x-covered SiO₂ surfaces are examined to investigate the initial nucleation process. This work will be important to reveal the mechanism and feasibility of atomic layer deposition of metals using N-plasma.

References
1. Maimaiti, Y.; Elliott, S. D., Precursor adsorption on copper surfaces as the first step during the deposition of copper: a density functional study with van der waals correction. J. Phys. Chem. C 2015, 119 (17), 9375-9385.

Under optimized conditions, atomic layer deposition (ALD) allows processing of films with high uniformity and conformity on 3D surfaces. These characteristics make ALD an appropriate tool to produce ultra-thin films for a variety of applications in microelectronics, optoelectronics, catalysis, renewable energy and more. Nonetheless, ALD processes still suffer from some non-uniformity of the deposited films on large area wafers, from initial substrate inhibited island growth and from the formation of an interfacial layer with uncontrollable characteristics. In order to overcome such drawbacks, the thorough understanding of the phenomena and mechanisms involved during the ALD process is essential. In this work, we present a framework for the integrated study of the ALD of Al₂O₃ from TMA and H₂O on Si, using a commercial ALD reactor.
The framework consists of a combined computational and experimental approach. A three-dimensional Computational Fluid Dynamics (CFD) model is built for the ALD reactor, designed to treat large area 20 cm substrates. The model aims to investigate the effect of the reactor geometry and process parameters on the gas flow and temperature fields, and on the species distribution on the heated substrate surface. We couple this model to a surface chemistry one, which considers adsorption, desorption and reactions on the substrate surface. We combine it with ellipsometry measurements to access the limiting mechanisms, the competition between surface phenomena and their effect on ALD growth. This integrated (surface chemistry/kinetics and CFD) approach reveals a direct link between film uniformity and deleterious transport phenomena due to gas flow recirculation and low temperature zones in the reactor.
We also develop a computational model based on geometric principles to study the initial nucleation and growth steps of the Al₂O₃ film. Using XRR measurements, we evidence a substrate inhibited growth regime, attributed to the low reactivity of the HF-cleaned Si surface, together with an island growth mode for Al₂O₃. Due to these phenomena, 25 ALD cycles are needed to produce a continuous film. Our computational model, numerically reproduces these findings and allows deriving the film nucleation mechanisms and the phenomena leading to island growth.
We call for a complete range of techniques, including TEM, XPS, EDX and ToF-SIMS to get insight to the morphology and chemical nature of the deposited film. We also precisely analyze the chemical composition of the (Al, O, Si) interfacial layer to get insight in the mechanisms of its formation. We show that Si oxidation occurs during the island growth, catalyzed by the presence of Al, while it is also fed by species interdiffusion through the film.
In order to enhance the initial surface reactivity, we implement an in situ plasma N₂-NH₃ pretreatment of the HF-cleaned Si substrate prior to ALD. We use STEM coupled to EDX to show that the initial deposition is clearly increased on the pretreated surfaces, resulting in a linear ALD regime even after 5 ALD cycles. Furthermore, a Si_xN_yH layer is formed by the N₂-NH₃ plasma pretreatment, which acts as a barrier layer, reducing the oxidation of the Si substrate beneath it.
This integrated analysis provides a general framework for the study of the mechanisms and phenomena involved during the ALD process as well as the non-ideal aspects leading to the drawbacks of ALD. This can lead to the determination of adequate surface pretreatments, able to combat the witnessed drawbacks during metal-oxide ALD on Si, thus paving the way for the effective production of nanometric thin films with sharp interfaces for microelectronic and other applications of the future.

Progress in thin-film deposition methods often depends on our understanding of the underlying elementary steps at the level of atomic and electronic structure. For key questions, theoretical investigations can help by complementing the experimental investigations. Nevertheless, the goal for theoretical approaches is to move beyond interpretation of already available data and provide predictions and guidance for future investigations. Predictive computations require high accuracy which directly leads to ab initio methods like density functional theory (DFT) or wave function-based methods combined with methods for larger length and time scales.
For CVD, the low-temperature growth of GaP on H/Si(001) by metal-organic vapor phase epitaxy (MOVPE) is taken as an example to highlight the strengths (and limitations) of these ab initio approaches. For the first step - the gas phase decomposition of precursors - we find very high barriers for several reactions proposed earlier for tert-butylphosphine (TBP) and triethylgallane (TEGa) and unusual β-H elimination reactions relevant for ligand design.[1] DFT approaches including methods for treating dispersion interactions (DFT-D) are in good agreement with gold-standard quantum chemical coupled-cluster methods.
This lends confidence for using these methods to reveal characteristics of surface adsorption for TBP – accurate enough to predict the outcome of growth experiments and ex-situ scanning tunneling microscopy measurements. Unusual electronic vacancy stabilization effects are revealed by theory to lead to non-statistical growth in the nucleation phase.[2]
In a joint endeavor of experiment and theory, we could further reveal the thermodynamic and kinetic reasons for the unexpected structure of GaP/Si interfaces. Intermixing models established previously are not sufficient to explain the pyramidal structure observed experimentally. Instead, relative facet stabilities and ad-atom mobility in the interface region needs to be quantified to find a rationale for the atomically-resolved measurements.[3] Further fruitful experiment-theory collaborations resulted in transfer of these ideas and approaches toward modelling of CVD of boron carbides.[4]
Recently, the insight gained from ab initio CVD modelling are transferred to ALD research. We will present initial results focusing on surface chemistry in ALD and AS-ALD of prototype systems. The specific angle taken is based on electronic structure analysis conducted with our recently proposed energy decomposition analysis for extended systems (pEDA) giving quantitative insight in surface-adsorbate bonding crucial for the growth mechanism.[5]

[1] a) A. Stegmüller, P. Rosenow, R. Tonner, PCCP, 16 (2014) 17018; b) A. Stegmüller, R. Tonner, Inorg. Chem. 54 (2015) 6363; c) A. Stegmüller, R. Tonner, Chem. Vap. Deposition 21 (2015) 161.
[2] A. Stegmüller, K. Werner, M. Reutzel, A. Beyer, P. Rosenow, U. Höfer, W. Stolz, K. Volz, M. Dürr, R. Tonner, Chem.--Eur. J. 22 (2016) 14920.
[3] A. Beyer, A. Stegmüller, J. O. Oelerich, K. Jandieri, K. Werner, G. Mette, W. Stolz, S. D. Baranovskii, R. Tonner, K. Volz, Chem. Mater. 28 (2016) 3265.
[4] a) M. Imam, K. Gaul, A. Stegmüller, C. Hoglund, J. Jensen, L. Hultman, J. Birch, R. Tonner, H. Pedersen, J. Mat. Chem. C 3 (2015) 10898; b) M. Imam, L. Souqui, J. Herritsch, A. Stegmüller, C. Hoglund, S. Schmidt, R. Hall-Wilton, H. Hogberg, J. Birch, R. Tonner, H. Pedersen, J. Phys. Chem. C 121 (2017) 26465.
[5] M. Raupach, R. Tonner, J. Chem. Phys. 122, (2015), 044707.

Whilst heavy chalcogenide containing materials play an important role in a number of opto and thermo electric materials, the development of single source precursors containing elements such as Se and Te has been hindered by the inherent weakness of the elemental bonds. Without the use of expensive techniques or phosphorus containing precursors, metal selenide deposition commonly results in poorly orientated, polyphasic thin films. ¹ Aerosol Assisted Chemical Vapour Deposition (AACVD) circumvents the need for volatile materials and uniquely allows access to kinetically stable phases which are typically inaccessible through high temperature deposition techniques. ²

Based on our work on the development of thioureide complexes for the AACVD of metal sulphide thin films, such as SnS and ZnS, we describe here our initial studies into the synthesis, characterisation and utilisation of the seleno analogues (selenoureides) for the deposition of SnSe and ZnSe thin films. ^{3, 4}

We have successfully deposited uniform, thin films of phase- pure SnSe which display temperature dependent orientation at temperatures as low as 200 °C. Highly orientated ZnSe polycrystalline thin films at low temperatures have also been produced utilizing the same ligand system.

[1] S. Mahboob, S. N. Malik, N. Haider, C. Q. Nguyen, M. A. Malik and P. O. Brien, J. Cryst. Growth, 2014, 39–48.
[2] C. E. Knapp and C. J. Carmalt, Chem. Soc. Rev., 2016, 1036–1064.
[3] I. Y. Ahmet, M. S. Hill, A. L. Johnson and L. M. Peter, Chem. Mater., 2015, 7680–7688.
[4] H. Sullivan, J. Parish, P. Thongchai, G. Kociok-Köhn, M. S. Hill and A.L. Johnson, Inorg. Chem., 2019, 2784–2797.

Vanadium dioxide (VO₂) is a strongly correlated oxide widely studied for applications in electronics due to its metal-insulator transition at approximately 68°C. While thin films and nanostructures of VO₂ have been grown on common rigid substrates like sapphire, TiO₂, Si and Ge, further expansion of the scope of their application lies in their growth on transferable or flexible substrates making it easier to integrate them into devices. In this work, we report the epitaxial growth of VO₂ nanowires on h-BN substrates by chemical vapor transport. We observe that nucleation and growth of VO₂ on h-BN proceeds very differently than that on c-plane sapphire: higher nucleation unexpectedly occurs on the h-BN surface and growth tends to proceed outward i.e. toward and beyond the edges of the h-BN substrate. We study the resulting epitaxial strain in h-BN by mapping a shift in a characteristic h-BN Raman peak. We further reveal that the temperature-driven phase transition of VO₂ on h-BN proceeds via a single phase boundary which is drastically different from that on substrates such as SiO₂ and sapphire, suggesting the presence of weak bonding at the VO₂/h-BN interface. Such growth of single crystal VO₂ nanostructures not only provides an ideal system to study properties of VO₂ based heterostructures but may also be extended to other flexible, transferable substrates.

Atomic layer deposition processes are fundamentally controlled by surface-gas phase reactions, and so depend to a great degree on the nature of the precursor. There are many general rules for the design of precursors: they should have strong metal-ligand bonds to prevent thermolysis, but these bonds should react to allow chemisorption of the precursor to the surface. The precursor should be low molecular weight, since (to ideal assumptions) vapour pressure is related to the inverse of molecular weight, and should be designed to minimise intermolecular interactions. The precursor needs steric protection to assist with self-limiting monolayer formation, but the metal centre must be accessible to reaction with both the surface and the second reactant. Many of these design rules are in opposition to each other, and so a balance must be struck.

When comparing families of precursors, a simple Figure of Merit has allowed us, in a straightforward manner, to compare and diagnose precursor performance. Using examples from across the periodic table, this presentation will demonstrate how a Figure of Merit can highlight when these design rules succeed and when they fail. For example, a molybdenum diimide dimer shows similar volatility to a variety of its base-stabilized monomers but has superior thermal stability. This ultimately makes the dimer a better precursor candidate. In this case, the coordination base in the monomers can be thermally reactive, leading to a low-temperature decomposition route. Similarly, a gold(I) precursor candidate shows its best performance with a ligand with an intermediate steric bulk, mainly due to stabilization of the precursor thermally.

The proper selection of tailored precursors plays a crucial role in chemical vapor deposition techniques for the fabrication of high-performance functional nanostructured films. Metal alkoxides are most suitable precursor class for depositing well-defined metal oxide thin films on various substrates with different surface chemistry and complexity, due to their high volatility and defined decomposition mechanism. We demonstrated the strong influence of the alkyl function in vanadium(IV) alkoxides on phase and morphology. Phase selective chemical vapor deposition of thermochromic vanadium dioxide was achieved by deploying homoleptic and monomeric [V(O^tBu)₄] as a precursor. The produced thin films of monoclinic VO₂ (M1) exhibited a small hysteresis at lower temperatures (63°) in the reversible thermochromic metal-to-semiconductor transition (MST), due to the anisotropic one-dimensional growth. In contrast, use of trimeric [V(OEt)₄]₃ generated conductive thin films of crystalline vanadium sesquioxide with a flower-like morphology. The in-situ reduction mechanism of the V(IV) precursor to V₂O₃ was elucidated by in-operando mass spectrometry. Moreover, novel volatile and stable heteroleptic bimetallic alkoxides ([LaFe(O^tBu)₄(L)₂] and [LaSn(O^tBu)₄(L)] with L = trifluoro-butenone-tert-butyl amine) as well organometallic noble metal complexes ([M^I(COD)(L)] COD = Cyclooctadiene; M^I = Rh, Ir) were synthesized by using tailored multidentate enaminolate ligand (L) systems. The interdependence of the organic backbone, choice of donor atoms and denticity in the ligand systems on the reactivity, stability and volatility of the resulting precursors were investigated by single crystal structure analysis, NMR-spectroscopy and thermogravimetric measurements (TG-DTA). Chemical vapor deposition of LaFe(O^tBu)₄(L)₂ and LaSn(O^tBu)₄(L) revealed the selective conversion into perovskite LaFeO₃ and pyrochlore Sn₂La₂O₇ thin films, respectively, which shows superior gas sensing characteristics in terms of sensitivity towards sulfur-containing gases (LaFeO₃) and selectivity towards hydrogen (Sn₂La₂O₇). The decoration of noble metals on nanostructured metal oxides was performed in a second CVD step by using our developed heteroleptic janus-type precursors [M^I(COD)(L)] enhanced their (photo-)catalytic as well gas sensing performance.

The ability to incorporate metal atoms into nanoporous materials such as metal organic frameworks (MOFs) in a well-controlled fashion provides new opportunities to prepare functionalized and modified materials for potential applications such as catalysis and gas separation. There are however new challenges that need to be overcome such as understanding the reaction mechanisms in order to develop structural and process optimization. MOFs possess three-dimensional structures, with complex pore architecture, leading to a number of possible processes (gas transport, adsorption and reaction) that are much more complex than on flat surfaces. To address these issues, we have combined in-situ infrared spectroscopy, X-ray-photoelectron spectroscopy and ab initio calculation to study the reaction of a number of common ALD precursors --trimethylaluminium (TMA), diethylzinc (DEZ), titanium tetrachloride (TiCl₄)-- with in several Zr-MOFs containing hydroxyl (OH) and water (H₂O) groups. Differentiating reaction with OH and H₂O groups is particularly interesting since their reactivity highly depends on both the chemical and structural (i.e. sterics) environments. We find that the OH groups in the Zr₆(μ₃-OH)₄(μ₃-O)₄(OH)_x(OH₂)_y cluster node do not all react at similar rates (i.e., the reaction pathway and energetics are highly dependent on their location, accessibility and chemical environment). For different OH-containing MOFs without H₂O groups, the activation temperatures for the TMA reaction with bridge OH of Zr₆ clusters decrease with their node connectivity, and are 250 °C, 150 °C and 24 °C for UiO-66-NH₂, Zr-abtc and MOF-808, respectively. Interestingly, the amine group in UiO-66-NH₂ is found to act as a catalytic active site by anchoring TMA molecules and facilitating their reaction with nearby hydroxyl groups, which is not observed in un-functionalized UiO-66. This synergistic effect between -NH₂ and -OH is fully elucidated by first-principles calculations. In addition, we find that TMA easily reacts with water adsorbed on the external surfaces of wet MOFs crystals at room temperature, forming a thick Al₂O₃ blocking layer on the periphery of MOFs crystals. These findings provide a basis for the design and synthesis of new MOFs structures requiring ALD for new applications.

Chemical vapor deposition (CVD) and surface-limited variant atomic layer deposition (ALD) are strongly dependent on the chemistry of the precursors that are used, affecting yield/throughput, stoichiometry, impurities and process temperature. In metalorganic precursors, the central question is therefore the choice of ligands that surround the metal center. Heteroleptic precursors (containing more than one type of ligand) are one way to compromise between conflicting chemical requirements, such as surface reactivity versus thermal stability in the gas-phase. A well-known example is the Zr precursor that combines cyclopentadienyl and amido ligands (ZyALD(TM), Air Liquide), used commercially in fabrication of DRAM memory for electronics.

However to date we have barely 'scratched the surface' of the vast chemical space of possible heteroleptic precursors. As an example to illustrate the magnitude of the design problem, 715 chemically-distinct heteroleptic complexes can be formed by combining four of 10 ligands around a tetravalent metal center, and 8,855 complexes can be formed from a library of 20 ligands. Clearly, an exhaustive experimental analysis is not possible. Instead we look to computational screening to narrow down the search to the most promising options. Here we present a computational approach for screening metal precursors with respect to a property that is crucial for CVD and ALD: thermal stability. The computational strategy is illustrated on the example of Zr precursors for zirconium nitride, a material that is deposited as a hard coating to protect industrial parts in corrosive environments.

CVD precursors should decompose unimolecularly at elevated temperature in the reactor and we therefore seek complexes with moderately low energies for homolytic bond dissociation (i.e. into a pair of radical fragments). ALD on the other hand requires the gas-phase stability of precursors to be as high as possible, as this dictates the upper temperature limit for ALD. Such stability should be balanced against the requirement of high reactivity of ligands towards the growing surface in ALD.

We first enumerate over ligands. Even a small ligand library of just six O-free, N-bearing ligands (amines, amidinates, cyano and guanidinate, with various alkyl groups) gives 6^4=1296 possible complexes. However, 94% of these can be neglected as symmetrically equivalent or over-coordinated, leaving 81 for optimization with density functional theory (DFT). To study thermal stability requires a second phase of enumeration, over the 1280 different bonds that can be broken in these complexes (612 C-C and C-N bonds, 446 C-H bonds and 222 metal-ligand connections), yielding a set of radical fragments that are also computed at the DFT level. It is clear that every step in this computational workflow requires robust and efficient automation.

DFT reveals that the lowest bond dissociation energies are obtained for cleaving intact ligands from the metal, ranging 133-471 kJ/mol (compared to higher energies for ligand decomposition by abstracting radicals: 198-371 kJ/mol for the ethyl radical, 233-401 kJ/mol for methyl and 310-463 kJ/mol for the hydrogen atom). The least thermally stable complexes are found to be zirconium amides with bidentate amidinates/guanidinates, and so we predict that these would be the best CVD precursors. We observe very little steric effect when replacing methyl groups with ethyl groups. By contrast, cyano groups have a stabilizing influence, suggesting that they would be useful as spectator ligands in heteroleptic precursors for ALD. On the basis of this sample system, we discuss the general requirements for chemical enumeration software and the limits faced by automation.

We investigate a new process for the low-temperature atomic layer deposition (ALD) of high-quality insulating layers based on alumina. In organic opto-electronics, alumina thin films play a major role as insulating material (gate dielectric), passivation layer or encapsulation layer (gas permeation barrier). Atomic layer deposition (ALD) is known for growing amorphous alumina with high density and high electrical quality, but the deposition conditions usually require either a temperature above 100°C or highly oxidative condition (oxygen plasma or ozone) to obtain a reasonable deposition rate and high material quality. A method for producing high quality alumina below 60°C using low oxidative conditions would be highly welcome for the highly sensitive materials used in organic electronics.
By incorporating zinc as dopant during the deposition of alumina, we demonstrate the ALD of high-quality alumina films at room-temperature, while keeping a single ALD cycle below 1 minute. The resulting material shows physical properties very close to alumina grown above 80°C. The films with 5% Zn content are amorphous, with a density above 3 g/cm³, and a dielectric constant of 7.7 ± 0.4. The gas permeation rate though the zinc-doped alumina thin films has also been investigated, showing gas barrier performances comparable to alumina deposited above 80°C. In order to understand the role of zinc doping in the deposition mechanism, we have performed in-situ mass measurements using a quartz crystal microbalance, and have analysed high resolution X-ray photoemission spectra. They reveal that zinc might help the growth of alumina at low temperature by avoiding trapping of water molecules on the surface.

Chemical Vapour Deposition (CVD) of Transition Metal Dichalcogenides (TMDs) is a promising fabrication technique that can enable both scalable synthesis and high crystal quality, both required for applications of TMDs in monolayer form. The state-of-the-art of CVD of WSe2 often relies on the use of large quantities of H2 gas as reducing agent for Se0, which represents a hazard in research laboratories and for the future industrial upscaling. Furthermore, metal-organic CVD (MOCVD) syntheses employ H2Se along with hydrogen gas, increasing the potential hazard of the process. Here, we present a H2-free low pressure CVD of WSe2 based on the chalcogen precursor chemistry. We discovered that ZnSe, containing already reduced Se (Se-2), allows the fabrication of large WSe2 monolayer flakes as well as thin films with grain size up to 100 µm. The uniqueness of this inorganic Se-precursor have been demonstrated by performing synthesis from Se powder and other inorganic Se compounds (Cu2Se, CdSe and Na2Se), which did not lead to evidence of WSe2. The chemical reaction steps have been identified and will be discussed. In addition, since our W- and Se-precursors possess similar volatilities we have simplified our CVD setup from a two-zone furnace, usually used for TMDs growth, to a single-zone furnace. The growth temperature has been selected according to vacuum TGA data, whereas the other growth parameters, i.e. pressure, growth time and quantities have been optimised to yield mono- and few-layered WSe2 crystals. The synthesized material possesses optical properties comparable to exfoliated crystals, assessed by room temperature (RT) and low temperature PL measurements on WSe2 deposited onto SiO2. The RT-PL peaks exhibit FWHM of ~50 meV, while low T PL shows a defect-bound exciton signal similar to mechanical exfoliated crystals. XPS analysis demonstrate high chemical purity of the synthesized material, and absence of Zn in the flakes.

Nanostructured advanced materials require challenging synthetic methods ultimately suited for large scale fabrication of materials with specifically designed properties. Vapor phase routes are appealing synthetic approaches following a bottom-up building strategy, but for the synthesis of multicomponent nanostructures and films, the availability of suitable precursors is of crucial importance.
The various vapor routes require precursors with different properties, thus in conventional Metalorganic Chemical Vapor Deposition (MOCVD), the metal sources must be thermally stable and volatile, liquid assisted MOCVD processes require soluble complexes and atomic or molecular layer deposition (ALD, MLD) routes need volatile and reactive sources.
With this perspective, the presentation will focus on metal adducts bearing linked β-diketonate (β-dik) and ancillary Lewis ligands L of general formula “M(β-dik)_nL”. The relationships involving the molecular architectures and mass transport properties of the adducts in function of the metal nature (M= alkaline, alkaline-earth and rare-earth metal) will be discussed. The effects due to the nature of the β-diketonate (1,1,1,5,5,5-hexafluoroacetylacetone or 2-thenoyltrifluoroacetone) moiety as well as of the ancillary ligand L [L= monoglyme (dimethoxyethane), diglyme (bis(2-methoxyethyl)ether), triglyme (2,5,8,11-tetraoxadodecane), tetraglyme (2,5,8,11,14-pentaoxapentadecane), tmeda (N,N,N’N’-tetramethylethylendiamine], on mentioned properties will be addressed as well.
The perspectives of applications of these precursors to MOCVD and MLD of advanced materials for energy harvesting and energy conversion are presented.
In regard to energy harvesting applications, novel lithium precursors will be described for the MOCVD of LiNbO₃ thin films. As energy conversion systems, alkaline earth fluorides such as MF₂ (M: Ca, Sr) and the multicomponent MREF₄ (M: Na,K; RE: Y,Gd) are studied as materials for the production of new and more efficient photovoltaic (PV) devices due to their efficient hosts behavior for upconversion (UC) or downconversion (DC) processes.
Finally, a case study of application of MLD to the fabrication of inorganic/metalorganic hybrid system will be commented.
Part of this work is supported by the European Community under the Horizon 2020 Programme in the form of the MSCA-ITN-2016 ENHANCE project, Grant Agreement N.722496.

Metal halide compounds tend to possess large bandgaps, optical transparency, ion conductivity, and semiconducting or dielectric properties. Thin films of metal halides are thus promising for use in a wide range of applications, from optical coatings and imaging devices to optoelectronic devices like photovoltaics. To be useful in many of these applications, the metal halide thin films must meet several requirements, including continuity, purity, and smoothness. Chemical vapor deposition (CVD) is one of the main techniques used in industry to fabricate device-quality films of other materials, because of its molecular-level control of the fabrication process. However, research efforts in the development of a CVD process to deposit continuous metal halide thin films have met several reactivity challenges. The few known metal halide CVD and ALD processes have been confined almost entirely to metal fluorides, and have typically required ancillary metal halides as the halide source. A more general route to metal halide vapor deposition, such as one using the hydrogen halides (HX) as the halide source, would be of considerable interest. Some researchers have succeeded in producing discontinuous “islands” of metal halides (e.g., CuCl, CuI) using HX as a precursor, but continuous thin films of metal chlorides, bromides, and iodides have remained elusive.

In response to this challenge, we have developed a pulsed chemical vapor deposition method that produces continuous CuBr thin films by reaction between HBr gas and vinyltrimethylsilane(hexafluoroacetylacetonato)copper(I). CuBr films were grown in a custom-built, hot-walled ALD reactor, at a range of substrate temperatures, between 65 and 135 °C. Rutherford backscattering (RBS) was used to quantify and confirm the 1:1 Cu:Br stoichiometry, and the RBS peak shapes indicate compositional uniformity throughout the film. γ-phase CuBr was confirmed by X-ray diffraction. Scanning electron microscopy was used to characterize the morphology of the films resulting from a variety of temperature and substrate conditions. We have found that the substrate surface can strongly influence film morphology, and that appropriate substrate choice can enable device-quality film continuity. Whereas CuBr islands were produced when grown on oxide surfaces such as SiO₂, we found that continuous CuBr films could be produced on other surfaces, such as glassy carbon, silicon nitride, or platinum.

To the best of our knowledge, these results constitute the first continuous, non-fluoride metal halide thin films deposited by CVD using the hydrogen halide as the vapor source. Our method provides a reaction pathway that may offer a more general route to CVD of metal halide thin films.

Transition metal sulfides, and in particular nickel sulfides with various compositions and phases, are gaining attention due to their potential application in practical devices.^1–3
We have previously demonstrated the bottom-up synthesis of bulk metal sulfides using atmospheric pressure chemical vapor deposition (AP-CVD), with elemental sulfur as a precursor.⁴ The method we developed allows us to synthesize a wide range of materials; by tuning precursors and process parameters, we can somehow control the morphology, stoichiometry, and structure of the synthesized materials.
Here, we present the synthesis of arrow-like structures of nickel sulfide using chemical vapor deposition. To the best of our knowledge, these arrow-like morphologies (with pyramidal termination) were not reported before. A carpet of long nickel sulfide arrows grew on a nickel foil. The arrows are up to 2 mm long with a high aspect ratio and a diameter of a few microns. XRD examination of the samples indicates Ni₃S₂ stoichiometry although detailed EDAX along the arrow indicates slight changes in stoichiometry as a function of the position in the arrow (closer or farther from the tip).
We will discuss the synthesis and the growth mechanism of these original structures based on extensive characterizations such as HRSEM, EDS, FIB, and HRTEM.




References:
1. Wei, W. et al.Partial Ion-Exchange of Nickel-Sulfide-Derived Electrodes for High Performance Supercapacitors. Chem. Mater.26,3418–3426 (2014).
2. Zang, X. et al.Template-Assisted Synthesis of Nickel Sulfide Nanowires: Tuning the Compositions for Supercapacitors with Improved Electrochemical Stability. ACS Appl. Mater. Interfaces8,24645–24651 (2016).
3. Sun, C. et al.Phase-controlled synthesis of α-NiS nanoparticles confined in carbon nanorods for High Performance Supercapacitors. Sci. Rep.4,7054 (2014).
4. Itzhak, A. et al.Digenite (Cu₉S₅): Layered p-Type Semiconductor Grown by Reactive Annealing of Copper. (2018). doi:10.1021/acs.chemmater.8b00191

Molecular layer deposition (MLD) is an increasingly popular process for the growth of organic and coordination polymer ultrathin films. This vapor-phase, layer-by-layer technique, which relies on the same principle of self-limiting surface reactions as atomic layer deposition, has shown promise for the development of many applications that require conformal organic-containing coatings. However, in many technologies, such as membranes, catalysts, and semiconductor devices, there is a growing need for ultrathin films with precisely controlled pore structure and crystallinity. Despite recent developments in MLD, there is still a significant gap in our understanding of the fundamental mechanisms that affect film structure, such as the origin of the film growth rate or their molecular-level properties. In this presentation, we present results of our recent studies to understand such mechanisms and discuss prospects for better controlling film structure moving forward.
First, we discuss our exploration of the growth behavior of MLD films by examining trends in film properties as a function of backbone flexibility [1]. Our results suggest that changes in growth rate between the most rigid and most flexible backbones (4 Å/cycle vs 1 Å/cycle) are not caused by differences in length of molecular precursors, chain orientation, or film density, but are instead caused by an increased frequency of terminations in the more flexible chemistries. These terminations likely result from monomers reacting with more than one functional group on the underlying surface, which reduces the total number of available reactive sites. We further elaborate on the relationship between the number of reactive sites and the film growth rate by modeling growth behavior after an intentional reduction in the number of reactive sites [2]. We show that terminations caused by dual-reacting monomers reduce the film growth rate; however, the adsorption of monomers likely reintroduces reactive sites, preventing the complete cessation of film growth. We also present diffraction and infrared absorption data, which suggest that films consist of a mixture of upward growing chains and horizontally aligned layers of paracrystalline polymer segments [1]. Combined, these results provide a clearer picture of the disordered nature of MLD and provide insight for the design of future ultrathin film synthesis chemistries. We conclude by highlighting recent work to create structured ultrathin films and propose several avenues of research to create films with precise morphological control.
1. D. S. Bergsman, et al., Chem. Mater, 2017, 29, 1192
2. D. S. Bergsman, et al., Chem. Mater, 2018, 30, 5087

Gallium oxide (Ga₂O₃) possesses an extremely wide bandgap around 5 eV and has attracted more attention for applications to power devices and optoelectronic devices. Ga₂O₃ exhibits five polymorphs of α, β, γ, δ, and ε phases. [1] Among these polymorphs, only metastable-phase orthorhombic ε-Ga₂O₃ exhibits large spontaneous polarization and ferroelectricity. [2, 3]

The polarization of ε-Ga₂O₃ is expected to lead to two-dimensional electron gas (2DEG) induced at ε-(Al_xGa_1-x)₂O₃/ε-Ga₂O₃ hetero-structure interface, like conventional AlGaN/GaN-type high electric mobility transistor (HEMT). Furthermore, the ferroelectricity of ε-Ga₂O₃ expected to be ferroelectric devices, such as ferroelectric field effect transistor (FeFET).

It has been reported an epitaxial growth of ε-Ga₂O₃ thin films by physical vapor deposition (PVD) and chemical vapor deposition (CVD) techniques. Among various growth techniques, we have focused on mist CVD for the epitaxial growth of ε-Ga₂O₃ thin films. Also, we previously reported that orthorhombic ε-(Al_xGa_1-x)₂O₃ and ε-(In_xGa_1-x)₂O₃ alloy films were epitaxially grown via mist CVD. [4, 5]

To achieve ε-Ga₂O₃ based device applications, such as HEMT and FeFET, high crystal quality ε-Ga₂O₃ thin films with smooth film surface are essential issues of the crystal growth. In this study, we report on investigations to improve crystal quality and to obtain a smooth surface using mist CVD. In the mist CVD, gallium (III) acetylacetonate is generally utilized as Ga precursor dissolved in a water solution. We have previously demonstrated that controlling the Ga/O ratio by varying the Ga precursor concentration in the solution enables smooth film surface under stoichiometric conditions [6], indicating that chemical reactions under the growth of ε-Ga₂O₃ film can be controlled in the mist CVD process like other CVD techniques.

Besides, we performed two approaches to improve the crystal quality of ε-Ga₂O₃ thin films using mist CVD; use of carbon-free gallium (III) chloride as CVD precursors and a surfactant effect of heavy atom addition in a precursor solution.

We obtained ε-Ga₂O₃ thin films grown on epitaxial substrates using gallium (III) chloride with higher crystal quality than that using gallium (III) acetylacetonate, considering incorporation of impurities, such as carbon was reduced.

The surfactant effect can be expected to enhance the migration of atoms on the film surface in the growth process. We utilized bismuth atom as a surfactant and the precursor was bismuth basic nitrate dissolved in the starting solution with gallium precursors. FWHMs of XRC for (004) ε-Ga₂O₃ with and without Bi precursors were estimated as 0.13 and 0.43, respectively, indicating that the crystal quality was drastically improved by only containing Bi precursors in the starting solution. Furthermore, the RMS roughness calculated from AFM was also improved 1.0 nm to 0.4 nm by using Bi precursors. Thus, we assumed that the Bi-assisted growth of ε-Ga₂O₃ thin films in the mist CVD method to improve the crystal quality and the surface morphology. We believe that our results in the mist CVD will be useful in the development of device applications.


[1] R. Roy et al., J. Am. Chem. Soc. 74, 719 (1952).
[2] F. Mezzadri et al., I Inorg. Chem. 55, 12079 (2016).
[3] M. B. Maccioni et al., Appl. Phys. Express 9, 041102 (2016).
[4] D. Tahara et al., Appl. Phys. Lett. 112, 152102 (2018).
[5] H. Nishinaka et al., CrystEngComm 20, 1882 (2018).
[6] D. Tahara et al., Jpn. J. Appl. Phys., Part 1 56, 078004 (2017).

Atomic layer deposition (ALD), self-limiting version of CVD, is a very attractive method due to superior characteristics of conformal deposition, precise thickness and composition control, and pinhole-free thin films. The advantages make ALD used widely in various applications such as microelectronics, photovoltaics, and energy storage systems. One of the most widely used and studied ALD materials is alumina which has great mechanical, optical, and electrical properties and functions as high-k dielectrics, chemical protection layers, and encapsulation layers for flexible devices. However, because of the brittle nature of the ALD alumina, cracks in the ALD alumina layer in flexible devices occur commonly by bending. Considering the mechanical problem, it is most important to measure precisely mechanical properties of the ALD alumina. Despite the importance, studies on intrinsic mechanical properties of the ALD alumina have been very insufficient. Previous works measured mechanical properties of the ALD alumina by using indirect methods such as nanoindentation, nanobeam bending, and bulge test, which are difficult to obtain accurate mechanical properties due to substrate effect or complicated calculation. In this study, we conducted tensile tests on ALD alumina thin films on the water surface without substrate and obtained stress-strain curves and intrinsic tensile properties of the ALD alumina itself directly and easily. Deposition temperatures of the ALD alumina for tensile tests were 80, 100, 150, 200, and 250 °C, and deposition temperature effect on the tensile properties were investigated. The tensile properties such as Young’s modulus and strength were increased with the increase of deposition temperature up to 200 °C, and then the tensile properties were reduced again. We proposed the defect generation mechanism related to impurities and hydroxyl groups, where the major defects in the ALD alumina thin film are changed, depending on deposition temperature. The defect generation mechanism was verified by the composition, roughness, and density of the ALD alumina measured by XPS, AFM, and XRR, respectively. The deposition temperature effect on tensile properties of ALD alumina was explained successfully by using the mechanism.

Layers of only a few atomic layers are often deposited by ALD, intending high control over the properties of the film. In these cases, the characterization of the early stages of film growth is crucial to optimize the process and quality of the film. For example, the initial thickness distribution before layer closure, created by the nucleation process, will often remain after the film is complete. To analyze these early stages of growth requires very surface sensitive analytical techniques with excellent detection limits.
Specifically for area selective deposition, the demand for characterization increases even further. The deposition process becomes more complex, involving etching steps to remove nucleation on blocked areas. In this case, means of characterization are required which determine the effects of etching steps on the growing film, possible contamination and the level of success of the blocking.
A technique specifically suited for these application is Low Energy Ion Scattering (LEIS). By scattering noble gas ions from the surface of the sample, the mass of the atoms in the outer atomic layer is determined non-destructively. Due to specific charge exchange processes, the peaks in the scattering spectrum correspond only to the outer atomic layer, making LEIS the most surface sensitive technique to determine the elemental composition of a surface.
In addition, information from deeper layers is available in two ways: First of all, features in the spectrum contain information about the first few nm of the sample – especially for heavier elements, the in-depth distribution can be determined non-destructively. For more complex systems or light elements, sputter depth profiling can be applied as well.
We will highlight the very special analytical possibilities of LEIS in the context of thin film analysis. Examples from industrial as well as academic collaborations will demonstrate typical analytical tasks. These include compositional analysis of the outer layer to determine residues of the precursor (e. g. fluorine), layer thickness evolution with varying process parameters, and core-shell nanoparticle analysis with respect to shell closure.

Off-stoichiometric copper chromium delafossites demonstrate the highest values of electric conductivity among the non-doped p-type transparent conducting oxides. Values of electrical conductivity beyond 100 S/cm are measured, this is only one order of magnitude lower than the actual n-type standard semiconductors. Values for carrier concentration beyond 10²¹/cm^-3 are determined for the as deposited films. However, a good understanding of the doping source and conduction mechanism is required in order to tailor the optolectronic properties towards values required in industrial applications. Controlled annealing was performed on the as deposited samples. A dramatic five order of magnitude decrease for the conductivity was measured, without observing any significant phase changes or alteration of average chemical environment. The thermal treatment proves itself as a useful tool for: first, tailoring the electric properties (such as electrical conductivity or Fermi level) ; second, investigating the source of the high level of doping on this material. High resolution Helium Ion Microscopy, Secondary Ion Mass Spectroscopy and Transmission Electron Microscopy and Positron Annihilation Spectroscopy were used to investigate the fine morphological and structural changes in Cu_0.66Cr_1.33O₂ upon annealing processes. The results indicate the chained copper vacancies as source of p-type doping and suggest furthermore that the changes in electrical conductivities within the off-stoichiometric copper-based delafossites are triggered by atomic rearrangements. This process, similar with an Oswald ripening mechanism, is based on the dissolution of the chains of vacancies into single Copper vacancies migrating to grain boundaries. A significant rearrangement of copper and chromium atoms after thermal treatments is revealed. The inhomogeneous distribution observed in the as-deposited high conductive thin films evolves to a uniform atomic dispersion within the annealed low conductive samples.

The production of complex oxides by CVS, for instance perovskites, requires multiple metal organic precursors at a steady mass flow to ensure a constant elemental composition.

One suitable method is flash evaporation by an infrared laser where a mixture of solid precursors is instantly sublimed. The resulting nonequilibrium conditions lead to instant sublimation of the irradiated spot on the precursor powder bed without decomposition. As a result, the composition of sublimed precursors corresponds to the composition of precursors in the solid mixture. Hence, flash evaporation allows a steady mass flow of precursors which exhibit significantly different volatilities [1].

In the presented work, a new flash evaporation system based on a marking laser is introduced that rapidly scans a focused infrared beam over a precursor powder bed. By focusing the beam, significantly higher energy densities are reached, compared to existing systems. Furthermore, the fast scanning speeds allow the use of spatially separated precursors. Thereby, the sublimation rate of each individual precursor is tuned precisely by the number of absorbed laser pulses. The process parameter space is discussed for the synthesis of LaFeO₃.

[1] M. Winterer, V. Srdic, R. Djenadic, A. Kompch, A. and T. E. Weirich, Chemical vapor synthesis of nanocrystalline perovskites using laser flash evaporation of low volatility solid precursors. Rec. Sci. Instr. 78, 123903 (2007) (10.1063/1.2821234).

Acknowledgements:
Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Projektnummer 388390466 – TRR 247.

Vapor deposition (VD) techniques have been highly successful in a wide variety of nanotechnological applications. However, films prepared by current VD methods are easily either highly continuous or have difficulties in functionalizing surfaces far out-of-sight, and therefore scaling them down to complex nanostructures can be challenging. It has been shown that with capillary condensation (CC) it is possible to coat and fill nanostructures selectively and controllably from the vapor phase.[1]

CC is a phenomenon of gas-to-liquid condensation inside nanocavities below the saturation point of the condensate. Because CC scales negatively with the surface feature size, it is possible to coat or fill selectively those surfaces and spaces which have been considered to be the hardest to reach, for example complex nanopores, ledges, and particle interstices.[1] First a reagent is allowed to reach a gas-liquid equilibrium at a predefined partial pressure. Then it is fixed into solid phase by a chemical reaction.

CC based methods are capable of bypassing and reversing many traditional problems related to VD, such as limited scalability in nanostructures, low control over spatial selectivity, sensitivity to gaseous inhibitors, and sometimes high process complexity and harsh conditions.[1] Due to the simplicity of CC, the process is easy to apply with no prior expertise in simple reactors which vary from a closed container with overpressure venting[2], to self-built reactor line for high process control with manual[1] or computer-assisted controls[3], or likely even established iCVD reactors[4] if the parameters were optimized below supersaturation. CC has been utilized for processes with polyacrylates and polystyrene on various substrates with UV initiation[1], silicon on carbon with thermal initiation[2] and silica on titania with chemical initiation[3].

In the presentation we explore principles and applications for CC based coating methods. We also show early results on how the fillings could be used as nanotemplates for inorganic material deposition during Vapor Phase Infiltration.[5] Studying the fundamentals of CC based methods could create a foundation for a novel VD paradigm for toposelective vapor deposition in nano- and micron-scaled structures. Additionally, CC based processes could be attractive also for industry due to their simplicity.
[1] V.A. Lovikka, M. Kemell, M. Vehkamäki, M. Leskelä, Mater. Horiz. 6 (2019), 1230-7.
[2] T. Masuda, N. Tatsuda, K. Yano, T. Shimoda, Sci. Rep. 6 (2016), 37689.
[3] S. Kim, S.H. Ehrman, Langmuir 23 (2007), 2497–2504.
[4] K. Ichiki, B. Altemus, A. Gildea, J. Faguet, Thin Solid Films 635 (2017), 23-6.
[5] C.Z. Leng, M.D. Losego, Mater. Horiz. 4 (2017), 747–71.

Advanced technologies in electronics, energy production, and energy storage require increasingly robust protection against destructive environmental conditions, such as high humidity, condensation, reactive gases and organic contaminants. At the same time, conventional encapsulants can interfere with performance, due to modified surface roughness or changes in optical, mechanical or thermal properties of the substrate.

Chemical vapor deposition of conformal films addresses many of the challenges with changes in substrate properties. The relatively low thickness and high conformality of vapor-deposited coatings allows for minimal alteration in substrate properties while ensuring a high level of protection.

GVD Corporation is developing a vapor-deposited, multilayer gas barrier coating for environmental protection specifically for flexible substrates. GVD’s approach combines its proprietary method for vapor deposition of polymers, called iCVD, and plasma enhanced chemical vapor deposition into a single, streamlined process. The coating takes advantage of fine thickness control to achieve the barrier properties of brittle materials in flexible films. To maximize the commercial potential of this technology in advanced applications, like elastomeric seal protection and solar cell encapsulation, GVD has optimized the process for barrier performance and film deposition rate. The impact of coating uniformity, chemical structure and multilayer stack design will be discussed.

Metal oxide films are a common component in modern electronic devices, from cell phones to photovoltaics. Atmospheric Pressure Spatial Atomic Layer Deposition (AP-SALD) is a novel technique for producing these films, retaining the high conformity and precise thickness control of conventional ALD, but without expensive vacuum chambers or time-consuming purge cycles. Where conventional ALD may take hours to produce a 100 nm Al₂O₃ film, our AP-SALD system can produce the same film in under twenty minutes. It is beneficial to couple AP-SALD with characterization tools that can quantitatively determine the properties of a film as it is being deposited in real time. Thus, the properties of the film can be optimized immediately via the deposition parameters such as precursor concentrations, deposition speed, reactor head height, or temperature. Using a combination of reflectance spectrometry and a novel flexible printed circuit board (PCB) our group was able to characterize how doped and undoped ZnO films change during depositions under various conditions. This allows for observation of film nucleation mechanisms, and determining the thickness, optical band gap, resistivity, refractive index, and uniformity of the films while they are forming. Notably, analytical models (Cauchy, Tauc-Lorentz, Drude-Lorentz) are used to determine the optical properties, with a feed-forward neural network providing initial guesses for those models.

Carbon nanotubes (CNTs) hold great theoretical promise as electrical conductors due to a combination of high conductivity, flexure tolerance, tensile strength, and a low thermal coefficient of resistivity (TCR). However, the resistive junctions between individual carbon nanotubes and bundles of CNTs, along with misalignment within bulk structures presents a unique challenge in the translation of the individual electrical conductivity to bulk applications. One strategy currently under investigation to alleviate these issues is to integrate CNTs with traditional metallic conductors yielding a hybrid material which combines the metal’s high conductivity with the low density and TCR of CNTs. Copper has traditionally been chosen to produce such composites, although the metal has limited interaction with the underlying carbon nanotubes.
In this study, site-selective chemical vapor deposition (CVD) is used to deposit nanometal interconnection seeds onto a porous, low-density (0.12 g/cm³, ~9 mg/m) CNT template. This site-specific method utilizes localized changes in the resistance within the bulk CNT template to preferentially deposit an organometallic precursor at thermally active sites. The deposited seeds act as interconnections sites with an electrodeposited copper metal overcoating. Group 10 elements like nickel, palladium and platinum exhibit more favorable wetting characteristics with the CNTs, allowing for more efficient interaction between the copper overcoat and the underlying CNTs. The deposited mass and distribution of seeds may be tuned through process parameters including precursor temperature, current density, time, etc., allowing for sparse depositions of seeds localized to hot spots (<20% w/w metal) as well as high densities of seeds leading to interconnection of the deposited nanometal network (>80% w/w metal). These nanometal seeds can be deposited throughout the carbon template in relatively short periods of time (~1 hour) compared to previous techniques such as electroplating (>10 hours). The morphology of depositions from various seeding metals have been investigated using scanning electron microscopy. In an example study with a low mass (~20% w/w) of deposited platinum seeds, alteration of electroplating rate allowed for the tuning of properties such as electrical conductivity, specific conductivity, density and TCR. Conductivity as high as 31.3 MS/m and TCR as low as 0.97 x 10^-3 K^-1 were achieved at 94.4% w/w total metal mass through modification of deposition and plating parameters. These values are increasingly competitive with traditional copper, which has an electrical conductivity of 58 MS/m and TCR of 3.83 x 10^-3 K^-1 at room temperature. Overall, the present results demonstrate the potential of site-specific CVD towards the enhanced nanometal interconnection of carbon conductors (NICCs).

With continuous CD scaling of semiconductor devices, surface and interface engineering plays an increasingly important role in semiconductor device fabrication, and atomic level surface treatment has been incorporated in thin film deposition, lithography and etch processes. Radicals are highly reactive yet do not damage materials surfaces, thus have become popular precursors in emerging atomic surface engineering applications.
In the present study, a high flux of methyl radicals is generated in a down-stream plasma setup in combination with optimized gas flow, pressure and plasma source power. Methyl radicals from the down-stream plasma source can be applied for treatment of semiconductor, dielectric and various metal surfaces. Surface methyl group coverage can be controlled by methyl radical exposure time and substrate temperature, as evident in surface infrared (IR) characterization and surface contact angle measurement.
Methyl groups on semiconductor and dielectric materials surfaces are stable upon air exposure. Sub-monolayer methyl group coverage can passivate silicon and silicon germanium surfaces, effectively eliminate native oxide growth, and thus alleviate issues with tedious queue-time control in semiconductor fabrication flows. In addition, surface methylation offers unique materials protection against chemical corrosion, including a complete stop of silicon loss in aqueous tetramethylammonium hydroxide (TMAH), a significant reduction of silicon oxide loss in aqueous HF and KOH, and a significant reduction of silicon nitride loss in aqueous H₃PO₄ solutions.
Low-k, silicon oxide and silicon nitride films are subject to carbon loss upon exposure to oxygen-based dry etch chemicals, etc. The surface methylation process can be adopted to restore carbon content in the surface region of low-k materials, as indicated by a recovery in X-ray photoelectron spectroscopy (XPS) carbon intensity and surface wetting angle.
Various metal materials are incorporated in semiconductor devices as work function, interconnect liner and conductor materials, respectively. Surface chemistry of hydrocarbons on different metal surfaces has been well studied [1]. Methyl radicals break down to carbon residues on reactive metal surfaces (e.g. Co). In comparison, surface methyl groups can be converted to volatile hydrocarbon products on clean metal surfaces with moderate reactivity (e.g. Cu), and thus contamination of the metal surfaces can be negligible in this methyl radical process.
Overall, a selective surface methylation process is feasible on patterned wafers with exposed dielectric and copper surfaces. The selective surface methylation process can be inserted into standard back-end-of-line (BEoL) dual damascene interconnect fabrication process flow, leading to a reduction in effective capacitance with no increase of Cu line and via resistance.
The selective surface methylation pretreatment process can be further extended to area-selective chemical vapor deposition (CVD) and ALD, particularly for selective dielectric materials deposition on metallic surfaces. In our exploratory studies, ZnO ALD is carried out on silicon oxide surfaces with and without surface methylation pretreatment, and film growth is characterized by surface IR, transmission electron microscopy (TEM), XPS, X-ray fluorescence (XRF) and atomic force microscopy (AFM). In absence of surface methylation treatment, dielectric films can be deposited readily on silicon oxide surfaces. After a methylation pretreatment, however, silicon oxide surface exhibits an incubation period in ZnO ALD cycles. We propose that the dielectric film deposition on silicon oxide surface can be further suppressed with a periodic surface methylation treatment. Subsequently, a novel selective ALD process can be developed to deposit dielectric thin film on top of Cu interconnect structures vs dielectric materials in BEoL integration scheme.
[1] Bent, B.E., Chem. Rev. 1996, 96, 1361-1390 1361

With continued downscaling of device dimensions, ultra-thin two dimensional (2D) semiconductors like WS₂ are considered as promising materials for future applications in nanoelectronics. At these nanoscale regimes, device fabrication with precise patterning of critical features is challenging using current top-down processing techniques. In this regard, area-selective atomic layer deposition (AS-ALD) has emerged as a promising candidate for bottom-up processing to address the complexities of nanopatterning. Till date, AS-ALD of metals¹ and dielectrics² have been successfully demonstrated. However, AS-ALD of 2D materials has remained elusive. In this contribution, we demonstrate area-selective deposition of 2D WS₂ nanolayers by using a three-step (ABC-type) plasma-enhanced ALD process.
AS-ALD of WS₂ was achieved by using acetylacetone (Hacac) inhibitor (A), bis(tertbutylimido)-bis(dimethylamido)-tungsten precursor (B), and H₂S plasma (C) pulses. This process resulted in immediate growth on SiO₂ while a significant nucleation delay was observed on Al₂O₃, as determined from in-situ spectroscopic ellipsometry and ex-situ X-ray photoelectron spectroscopy measurements. The surface chemistry of this selective process was analysed by in-situ Fourier transform infrared spectroscopy. The analyses revealed that the inhibitor adsorbed on the Al₂O₃ surface, blocking precursor adsorption, while little or no inhibitor adsorption was detected on the SiO₂ surface where WS₂ was readily deposited. Furthermore, the area-selective growth was demonstrated on SiO₂ samples with patterned Al₂O₃ on top.
To improve the crystallinity, the AS-ALD WS₂ films were annealed at temperatures within the thermal budget of industrial semiconductor processing (≤ 450°C). The annealed films exhibited sharp Raman peaks, which is a fingerprint of highly crystalline WS₂. Furthermore, Raman line scans over the patterns showed very sharp peak intensity transitions at the SiO₂-Al₂O₃ interface which confirmed that annealing had no impact on selectivity.
To summarize, this work pioneered the combination of two key avenues in atomic-scale processing: area-selective growth and ALD of 2D materials. It is expected that the results of this work will lay the foundation for area-selective ALD of other 2D materials.

In area-selective atomic layer deposition (AS-ALD), the chemisorption of a precursor and reactant can be promoted or suppressed depending on the terminal groups of the substrate. For instance, nucleation and growth of an ALD film will be suppressed when the surface reaction sites are fully terminated with -CH₃ or -CF₃ groups, which have low reactivity with precursors or reactants. These chemical species that hinder the reaction on the substrate surface are called inhibitors, among which self-assembled monolayers (SAMs) have been widely studied because of their excellent inhibitor characteristics. However, the formation of a SAM on the surface of a substrate with excellent packing density requires extremely long times, which renders it not applicable to large-scale fabrication processes. Furthermore, chemisorption of a SAM requires the use of a specific substrate depending on the SAM head group, and carbon- or fluorine-containing byproducts with deleterious effects on devices can be produced during post-processing. Thus, the development of a simple approach for the fabrication of inhibitors without carbon and fluorine is required.
Hydrogen-terminated species are alternative candidates as inhibitors. The adsorption of a precursor or reactant can be inhibited when the substrate surface contains H-terminated reaction sites, which are formed by H₂ plasma treatment or hydrofluoric acid dipping. However, these processes may induce damage on the substrate or device, and complete passivation of the surface reaction sites is not ensured. Moreover, H-terminated species has poor stability and can be easily oxidized in ambient condition after only a few hours of exposure, loosing its inhibitor characteristic.
In this study, we firstly report the use of atomic layer-deposited SiO_x as an inhibitor for AS-ALD. Unlike the traditional hydrophilic SiO₂, the SiO_x formed by ALD using an aminodisilane precursor i.e., 1,2-bis(diisopropylamino)disilane, showed excellent hydrophobicity, even though the film was only one nanometer thick. From a thorough analysis of the film, including X-ray photoelectron spectroscopy (XPS) and Fourier-transform infrared spectroscopy (FT-IR), we confirmed that this peculiar hydrophobicity is caused by the presence of SiH_x species, not by surface carbon from incomplete ALD reaction or by hydrocarbon from the air. The excellent stability of the SiO_x film was also demonstrated. Thus, it retained its hydrophobicity over a month and after exposure up to 350 °C in dry gas or at 250 °C in the ambient atmosphere. The ALD growth inhibition was confirmed by comparing the deposition of ALD Pt on the bare substrate and on the SiO_x-deposited substrate; the nucleation of Pt starts on the bare Si substrate after the 100th cycle of ALD, whereas no Pt growth was observed on the SiO_x-deposited substrate. On the SiH_x-terminated surface, the suppression of chemisorption was confirmed by using first-principle calculation. Although the ALD-based inhibitor covers all the substrate regardless of the material present on its surface, the reaction sites can be generated by the combination with a functional metal oxide, TiO₂. Using the photocatalytic effect of TiO₂, the inhibited sites can be easily changed to reactive sites, enabling the subsequent chemisorption of a precursor/reactant. From these results, we are certain that an ALD-based inhibitor deposition technique can be a promising method for the next generation of AS-ALD.

To form uniform and conformal polymer thin films, particularly with nanometer resolution, chemical vapor deposition (CVD) is an attractive approach. Specifically, initiated and oxidative chemical vapor deposition (iCVD, oCVD) are widely used to cleanly deposit a variety of polymers. Polymers – such as acrylates, methacrylates, vinyls, ethers, silicones, and fluorocarbons – are cleanly deposited by iCVD through an array of heated filaments that thermally activates an initiator for polymerization. Conjugated (conducting) polymers – like thiophenes, pyrroles, and anilines – are easily deposited by oCVD using an oxidizing agent that oxidizes the monomer to create polymerizable species.

This talk will focus on our recent efforts in using iCVD and oCVD to apply thin polymer films on heterogeneous/textured (physical or chemical) substrates. Physically heterogeneous substrates, e.g. nanoporous particles, networks and membranes, are commonly found in gas separation, biomolecular sieving, sensors, water/oil repellency, catalysis, energy capture, and energy storage. Likewise, chemically heterogeneous substrates, i.e. surface chemical patterns or chemically disparate surfaces, are common in many areas, including integrated circuits, sensors, biology, and 3D batteries. Our work aims to fundamentally understand the chemistry and physics of iCVD and oCVD that would allow us to precisely control polymer growth on physically and chemically heterogeneous substrates. Such working knowledge enables us to apply these CVD techniques in energy storage and area-selective applications, which will be highlighted.

Circuit boards and integrated circuits require protection from both RF interference as well as against the physical probing used in reverse engineering. Typical conformal coatings such as urethane or parylene have been used for environmental protection of circuit boards, but are too thick to be applied at the wafer level. In addition, the standard coatings are purely dielectric in nature and do not have any conductive aspect which could be used for either anti-tamper or RF shielding applications.
This paper describes a multi-layer coating which combines both dielectric and conductive characteristics. The dielectric material is a highly cross-linked polysiloxane polymer fabricated using a hot-filament CVD process referred to as iCVD, or “initiated chemical vapor deposition”. The iCVD process was invented at MIT by Prof. Karen Gleason and accommodates a wide range of off-the-shelf monomers and precursors. In iCVD, a gas or mixture of gases is introduced into a reactor under mild vacuum in the vicinity of an array of heated filament wires. The gas decomposes into reactive species – radicals – that serve as monomer units. These units then migrate to a cold surface (e.g., the surface of an electronic device) on which they combine and grow into a conformal polymer thin film. That is, the monomer units successively add to one another, forming the dielectric polymer.
The conductive film within the multilayer coating is pure poly(ethylene-3,4-dioxythiophene) polymer [PEDOT] deposited using a coating process referred to as oxidative chemical vapor deposition (oCVD). The oCVD method combines the clean, solventless character of chemical vapor deposition (CVD) with the versatility of radical-based chemistry techniques used in bulk polymer processing. This deposition process produces pure, mechanically-flexible coatings of intrinsically-conductive PEDOT; these PEDOT coatings are conformal to the topology of the substrate and renders surfaces electrically conductive. The novelty of the oCVD process is in the highly-conformal deposition of the polymers, which show conductivities up to 10³ S/cm. oCVD coatings are suitable for conformal coverage of even the most complicated surface topologies, including foams, fibers, and fabrics.
The ‘as-deposited’ multilayer coating can provide protection against RF interference. However, if the dielectric and PEDOT layers are selectively patterned then the coating can be used as an active anti-tamper coating. In this paper, the deposition processes are described in detail. The properties of the individual films are presented, along with the results of multilayer and patterning tests.

With its superior physical properties and band structure, the graphene has become one of the most attractive materials of recent years. Among many graphene synthesis techniques, CVD is considered as the most important and efficient method to synthesis high quality and large area graphene for high-tech applications. In CVD-grown graphene is typically produced by decomposing a hydrocarbon gas on the surface of a transition metal catalyst. However, synthesized graphene must be transferred from the catalyst surface to desired surfaces for final product application. In the typical CVD-grown graphene transfer process, firstly, graphene on the catalyst surface is coated with a support layer to prevent disintegration of as-deposited graphene layers. After that, graphene is separated by selective etching of the catalyst using solvents. Usually, rigid materials such as silicon and glass are used as substrate. Recently, paper has an important place as a substrate in high-tech applications as compared with rigid substrates due to its lightweight, flexible, biocompatible and foldable structure. However, in the transfer of graphene on the paper, the presence of solvents can lead to damages on paper surface. In this study, paper surface was functionalized with hydrophobic thin films using initiated chemical vapor desposition (iCVD) to make an graphene transfer feasible. For this purpose, two different fluoropolymers, namely poly(hexafluorobutyl acrylate) (PHFBA) and poly(perfluorodecy acrylate) (PPFDA) were deposited on the paper surface by iCVD. Graphene layers were transferred on the encapsulated papers and their electrical conductivity performances were tested. When PHFBA film was used as a prime layer, the better conductivity performance was obtained from graphene-on-paper.Our results showed that, PHFBA thin film is more suitable for graphene transfer due to its very smooth surface structure. As-transferred graphene layer on PHFBA coated paper surface showed high conductivity, even after repeated folding and flattening cycles.

Initiated chemical vapor deposition (iCVD) enables high-purity polymer coatings at room temperature through a reactor design that segregates the high temperature initiation (e.g. T>200°C) and surface polymerization reactions. A complex interplay between mass transport, thermal management, and surface adsorption under vacuum drives continuous flow iCVD reactor design. This talk reviews a simplified evaporative iCVD reactor that can be assembled with off-the-shelf parts in a chemical fume hood. Aspects of steady transport and reaction rates are compromised for ease-of-use and low cost. However, the simple evaporative iCVD coater is capable of providing coatings suitable for testing and for some applications, e.g. gluing small parts together with submicron glue thicknesses. Along with the evaporative iCVD reactor process design, I detail a case study where it is used to achieve sub-micron thick glue coatings on laser compression target materials (Cu, Si, CH, Al, Mo, LiF, quartz). These glued foil stacks are compressed or shocked by a laser impact in order to probe materials science at extreme pressure and/or strain rate. Sub-micron glue thicknesses are particularly important in the design and interpretation of these modern high energy density materials science experiments because the experimental foils are thin (O[10 µm]) and thick glue layers can seed instabilities and shock reflections.

With initiated Chemical Vapor Deposition (iCVD), it is possible to deposit smart hydrogel thin films conformally into 3D-nanostructures for sensor applications. In this contribution, we report on the thin film synthesis of such a thermo-responsive co-polymer by iCVD for the first time: namely, poly(N-vinylcaprolactam) cross-linked with di(ethylene glycol) divinyl ether; short p(NVCL-co-DEGDVE). In water, the transition of the polymeric system between a swollen state below to a shrunken state above the corresponding lower critical solution temperature (LCST) is investigated by spectroscopic ellipsometry (SE). By water contact angle (WCA) measurements and nano-indentation experiments, we show that the transition is accompanied by a change in wettability and elastic modulus. As previously shown for other polymers, the amount of cross-linking is successfully used to tune the temperature-responsive behavior of the deposited pNVCL-based systems. Higher swelling and LCST, higher surface rearrangement and lower stiffness are achieved in less cross-linked polymers. Interestingly, pNVCL is also reported to show decreased transition temperatures for higher molecular weight systems. Thus, by changing the filament temperature during iCVD, it is possible to lower the LCST by almost 20°C, without changing the (nominal) composition and, thus, the maximum swelling. Overall, maximum swelling degrees of the polymer below the transition temperature of up to 250% of the dry thickness are achieved and the LCST can be tuned in the range of 16-40°C. For probing the applicability in sensor setups, these polymers are also investigated in terms of swelling in humid environment (relative humidity, RH). There, three regions could be identified: First, in rather dry environment (up to ~40% RH), the material responds by mainly filling open porosity, but not showing a temperature-responsive behavior. Second, up to ~80% RH, the response in swelling is close to linear to the measured RH. Third, in very humid environment, the swelling is highly non-linear and temperature-dependent. Moreover, the film thickness approaches the value that can be observed when the polymer is immersed in water at the respective temperature. Furthermore, the response of the polymer in water as well as in humid environment is observed to be very fast; e.g., it responds faster than the commercial sensor used for monitoring the RH in the measurement cell (8 s response time). Together with the biocompatibility reported for pNVCL, the knobs of filament-temperature and cross-linking to tune the described features of the temperature-responsive swelling behavior of these systems make them highly promising for biomedical and/or environmental (sensor) applications.

Following the current development of strategies for the fabrication of structured surfaces, we present in this work an innovative approach for the synthesis of thin films with tunable morphology. Our method is based on the controlled generation of surfaces instabilities in bilayer systems formed by a mechanically responsive plasma polymer films (PPF) synthesized by the Plasma Enhanced Chemical Vapor Deposition (PECVD) method in combination with stiffer coatings.

As a case study, PPF were grown from propanethiol on silicon or flexible polyethylene terephthalate (PET) substrates. AFM data (i.e. peak force quantitative nanomechanical property mapping, scratching experiment) reveals that the nature of the PPF is dramatically affected by the substrate temperature (T_s): from a high viscous liquid (η ~ 10⁶ Pa.s.) to a viscoelastic (E ~ 0,1 GPa) and finally to a stiffer elastic solid (E ~ 0,9 GPa) material when increasing T_s from 10°C to 45°C. This evolution in the mechanical properties is correlated with a pronounced increase in the cross-linking degree of PPF evaluated by ToF-SIMS measurements, combined with a statistical treatment of the data. This trend is ascribed to an increae with T_s in the flux of energy brought to the growing film by positive ions and normalised with respect to the total amount of matter deposited.

In order to inducing a morphological reorganization of the material, we have deposited an Al thin film (50 nm) by the magnetron sputtering technique on the top of a mechanically responsive PPF. The mismatch between the mechanical properties of both layers results in the spontaneous formation of a wrinkled surface. By tuning the thickness as well as the mechanical properties of the PPF layers, the height (i.e. from 0.4 to 5.2 µm) and the width (i.e. from 0.6 µm to 6.5 µm) of the nano/micro wrinkles can be easily tailored in a wide range offering a great flexibility in term of surface engineering.

The same methodology was applied on flexible PET substrates. To evaluate the potential of the Al-based wrinkled/PET material as a flexible thin-film electrode, the mechanical stability of the electrode is explored under severe and repeated deformations. For this purpose, applying a strain of 1%, the electrodes are bent up for 10,000 times. The electrical resistance, measured after each bending cycle, shows a relatively stable behavior (the resistance increase only by 15 %). In contrast, for a flat Al thin film directly deposited on flexible PET, the electrical resistance dramatically increases by 12,000 % after 500 cycles indicating failure of the electrode.

Our results reveal the attractiveness of our method for the fabrication of micro/nano pattern with tuneable dimensions with potential applications in flexible electronics.

iCVD (initiated chemical vapor deposition) is a powerful tool to deposit various kinds of functional polymer films. Especially, the process is capable of exquisite control of its surface composition, elastic modulus, and functions through the introduction of various combinations of functionalities onto target substrates. Compared to conventional liquid phase-based methods, the solvent-free process can generate polymer coatings of extremely high purity, which is essential for various field applications, because such residual impurities produce harmful side effects. In addition, the thin iCVD polymer films remained durable and robust via proper crosslinking, rendering them suitable for the applications in hard environments. Such advantageous characteristics of the iCVD process have been exploited for tissue regeneration, flexible/wearable electronics, biosensors, and functional surface modification for various devices. In this presentation, the controlling of surface properties of functional polymer films and their application to various fields will be reviewed.

A novel manufacturing method for the scalable, one-step synthesis of a Pt/C electrocatalyst by inductively coupled plasma-enhanced chemical vapor deposition is presented. The metal-organic precursor platinum acetylacetonate is used to deposit platinum nanoparticles (NPs) and a support, carbon nanowalls (CNWs), simultaneously in a single, uncatalyzed process at remarkably low temperatures of 350°C. In general, CNWs exhibit exceptional thermal as well as electrical conductivity and widely adjustable surface area, making them an ideal support for use in energy conversion applications. By adjusting specific process parameters, such as pressure, gas flow rate and temperature, the density, thickness and height of the CNWs, as well as the platinum loading and oxidation state of the resulting catalyst can be tuned according to requirements. Scanning electron microscopy and Raman spectroscopy are used to determine the CNWs’ structural and electronic properties, respectively. Additionally, X-ray photoelectron spectroscopy and Auger electron spectroscopy are used to determine the overall and spatially resolved chemical composition and oxidation state of the catalyst. High-resolution transmission electron microscopy (TEM) reveals the presence of homogeneously distributed platinum particles with a mean particle diameter below 3 nm, resulting in an exceptionally narrow particle distribution with a polydispersity index below 0.1. Furthermore, due to the simultaneous synthesis of both NPs and support, the nanoparticles are expected to be incorporated into the carbon matrix, which is supported by TEM tomography. Such incorporation would greatly improve the immobilization of NPs on the support and thus the long-term stability of the catalyst. Finally, due to the use of a metal-organic precursor, the process can be adjusted to deposit almost any transition metal/CNW-hybrid material, making it a versatile method for a variety of different applications.

Vapor phase infiltration (VPI) is a relatively new processing technique used for transforming polymers into organic-inorganic hybrid materials. VPI has been used to improve polymer mechanical properties, protect fabrics from UV and thermal degradation, dope conducting polymers, and act as a contrasting agent in electron microscopy for imaging phases of polymer blends. Recently, our group has explored a new application for VPI, the protection of thermoplastic polymers from solvent dissolution. In this study, poly(methyl methacrylate) (PMMA) thin films were infiltrated with trimethylaluminum (TMA) and water at different temperatures and to different depths of infiltration. The resultant AlO_x / PMMA hybrid films were then exposed to a variety of solvents to explore their stability. Chemical stability was found to vary non-linearly, with infiltration temperature. Films infiltrated at lower temperatures (70^oC and 100^oC) swelled or partially dissolved in good solvents for neat PMMA, such as toluene or chloroform, and partially dissolved in isopropanol and water, which are not good solvents for PMMA. In comparison, films infiltrated at higher temperatures (130^oC) showed enhanced solvent stability in most solvents, even those that dissolved neat PMMA. The increased solvent resistance is likely due to crosslinking between PMMA functional groups and TMA molecules, a reaction that has been reported to vary with temperature. Due to this variability, PMMA films infiltrated at low temperatures are only partially crosslinked while those infiltrated at high temperatures are fully crosslinked, making them more solvent resistant. The increased dissolution of hybrid films in certain alcohols and polar solvents is hypothesized to result from an interaction between the inorganic crosslinker and the solvent. We also found that complete transformation of the polymer into hybrid material was unnecessary for dissolution resistance at higher temperatures. An infiltration depth of 0.5 mm was sufficient for complete resistance to toluene dissolution at room temperature. For proof-of-concept, we applied this treatment to a quarter inch thick laser-etched PMMA sheet and then exposed it to toluene at 60^oC for 30 minutes. While the design on the neat PMMA version rapidly dissolved, the sheet with a 0.5 mm AlOx / PMMA subsurface layer showed nearly complete retention of its design. In this talk, we will explore these findings and discuss the differences in solvent stability of AlOx / PMMA hybrid materials as a function of temperature as well as investigate the underlying chemical and structural variations that yielded these results.

The combined ALD/MLD (atomic/molecular layer deposition) technique is strongly emerging as a viable technology for the fabrication of new types of functional metal-organic materials and inorganic-organic interfaces not readily accessible through any other fabrication route. Most excitingly, it is capable of yielding in-situ crystalline thin films of structures similar to those known for so-called coordination polymer or metal-organic framework (MOF) materials synthesized from solutions,^[1] or even entirely new compositions/structures.^[2] An attractive example is the in-situ lithiated and crystalline thin films of lithium quinone, a promising cathode material for a Li-ion microbattery;^[3] the lithium quinone structure contains the Li⁺ cations in a coordinatively unsaturated three-fold coordination, which explains why it can not be synthesized through conventional solution synthesis. Equally exciting are the new metal-organic structures based on photoactive azobenzene,^[4] or upconverting lanthanide-ions.^[5] Another interesting approach enabled by ALD/MLD is to build artificial superlattices or gradient materials with precisely tailored inorganic-organic interface frequencies;^[6] in such layer-engineered hybrid materials the periodically introduced monomolecular organic layers between nm-scale metal oxide layers may efficiently block phonon conduction, enhance mechanical flexibility, control band-gap or magnetic properties, etc.

References
[1] E. Ahvenniemi, M. Karppinen, Chem. Commun. 52 (2016) 1139.
[2] J. Penttinen, M. Nisula, M. Karppinen, Chem. Eur. J. 23 (2017) 18225; Chem. Eur. J. (2019) in press.
[3] M. Nisula, J. Linnera, A.J. Karttunen, M. Karppinen, Chem. Eur. J., 23 (2017) 2988; M. Nisula, M. Karppinen, J. Mater. Chem. A 6 (2018) 7027.
[4] A. Khayyami, M. Karppinen, Chem. Mater. 30 (2018) 5904; A. Khayyami, A. Philip, M. Karppinen, submitted (2019).
[5] Z. Giedraityte, M. Tuomisto, M. Lastusaari, M. Karppinen, ACS Appl. Mater. Interfaces 10 (2018) 8845.
[6] F. Krahl, A. Giri, J.A. Tomko, T. Tynell, P.E. Hopkins, M. Karppinen, Adv. Mater. Interfaces 5 (2018) 1701692.

Two-dimensional hexagonal boron nitride (2D h-BN) is a single - atom thick layer (monolayer) of alternating boron and nitrogen atoms, which is of great interest and potential due to its excellent electrical, optical, and mechanical properties and isostructural to graphene. To fully utilize the potential of 2D h-BN, wide area processing of high-quality 2D h-BN is of utmost importance. As of now, chemical vapor deposition (CVD) is the most promising method to synthesize 2D h-BN because it provides great control on the thickness of the deposited films (number of layers) and is the best candidate for the industrial scale-up process. In this study, a CVD system was designed for synthesis of two-dimensional hexagonal boron nitride with low and atmospheric pressure capabilities as well as a solid and liquid precursor delivery system. The customized CVD system is also capable of synthesizing other 2D materials like graphene and molybdenum disulfide since it can incorporate different types of precursors and reaction gases. This presentation explores the CVD aspects of h-BN synthesis with a specific focus on the designed CVD reactor system and associated components. h-BN was synthesized directly on non-catalytic silicon substrates via LPCVD, which eliminated the need for a transfer process that can introduce impurities and surface defects (wrinkles, folding). Subsequently, the effect of the precursor flow behavior on the growth of hBN was studied. Scanning electron microscopy (SEM), Raman spectroscopy, X-ray Photoelectron Spectroscopy (XPS), and UV-vis spectroscopy were then used to characterize the deposited h-BN and study the effect of precursor flow kinetics. XPS analysis confirmed the presence of h-BN. Raman spectroscopy further corroborated the formation of h-BN. The approximate crystallite size of the deposited h-BN was also determined by Raman spectroscopy analysis. SEM confirmed the crystallite size estimated from Raman spectroscopy analysis of between 60 – 100 nm.

Area-Selective Atomic Layer Deposition (AS-ALD) of aluminum oxide (Al₂O₃), driven by arrays of microcavity plasmas (MALD), has been demonstrated. Al₂O₃ films were grown at 300 K by dissociating oxygen in an array of microcavity plasmas. Metal-oxide-semiconductor capacitors (MOSCAP), fabricated from Al₂O₃ films deposited on p-Si by MALD with Al top ohmic contacts and Au bottom electrodes, exhibited a breakdown electric field strength of E_BR = 6.1 ± 0.2 MV/cm and a hysteresis width of < 1 mV, indicating the presence of negligible traps in the film. The dielectric constant of Al₂O₃ films, ε_ox, was measured from MOSCAP C-V curves to be ε_ox = 9.7 ± 0.3. The measured values for E_BR and ε_ox are higher than conventional values, exhibiting that MALD-grown films are suitable for MOS devices operating under demanding ambient conditions which require extraordinary radiation resistance, thermal conductivity, and chemical stability. The ability to grow Al₂O₃ films of high quality at 300 K prompted experiments to fabricate Al₂O₃ arrays of small feature size. AS-MALD of Al₂O₃ films was accomplished by lift-off lithography. With this process, Al₂O₃ films having a lateral dimension of 1 - 10 µm and a thickness of ~ 68 nm were deposited. Due to the complete reaction between precursors, the oxygen and aluminum stoichiometric ratio for amorphous Al₂O₃ films has been measured by EDX and RBS to be ~ 1.5 ± 0.1, demonstrating that negligible levels of impurities and oxygen vacancies exist in the films. All experimental results show that MALD process can be compatible with semiconductor area-selective patterning with narrow lateral dimensions through simple lithography and lift-off.

Corundum structured III-IV oxide materials have been attracted considerable interest from oxide based wide bandgap semiconductor researchers. Typical corundum structured α-Al₂O₃ are widely used as substrates for wide band gap semiconductors, such as GaN, ZnO, and α-Ga₂O₃. The α-Ga₂O₃ in the series of corundum structure exhibits bandgap energy of 5.3 eV and controlling carrier concentration by n-type doping. Further, the α-Ga2O3 produces the possibility to tune a wide range of bandgap by alloying with α-Al₂O₃ (8.8eV) and rh-In₂O₃ (3.7eV) exhibiting the same corundum structure. The epitaxial growth techniques are usually used for the synthesis of these materials. For example, the epitaxial growth supported by α-Fe₂O₃ buffer layers showing the same corundum allowed growing single phase rh-In₂O₃ thin films which have been difficult to synthesis [1].
In this study, the mist CVD method was utilized for the epitaxial corundum oxide thin films and solution mist dissolving precursors was used as CVD precursors. Any precursor can be used as mist CVD precursors that only dissolves the precursor in any solvent.Taking advantages of its feature, we have demonstrated that the phase controlling of corundum oxide thin films were successfully obtained by using various corundum substrates and α-Fe₂O₃ corundum buffer layers.
For the α-Ga₂O₃ epitaxial growth, most researchers have utilized only α-Al₂O₃ substrates with large lattice mismatch for α-Ga₂O₃. However, we featured the LiTaO₃ and LiNbO₃ substrates showing nearly corundum structures with smaller lattice mismatches for α-Ga₂O₃. XRD results revealed α-Ga₂O₃ epitaxial thin films were successfully grown on LiTaO₃ and LiNbO₃ substrates with α-Fe₂O₃ buffer layers. The epitaxial ε-Ga₂O₃ thin films were grown on LiTaO₃ and LiNbO₃ without buffer layer. This is because ε-Ga₂O₃ also has a similar oxygen atom arrangement of the corundum structure and the phase is more stable thermodynamically than that of α-Ga₂O₃. Furthermore, to compare the crystal quality of α-Ga₂O₃ layers on LiNbO₃, LiTaO₃ and α-Al₂O₃, we investigated XRC of asymmetric plane (10-14) α-Ga₂O₃. The FWHMs of XRC for α-Ga₂O₃ on LiTaO₃ and LiNbO₃ were evaluated to be 0.51 degree and that for α-Ga₂O₃ on α-Al₂O₃ is evaluated to be 0.85 degree, indicating that the LiTaO₃ and LiNbO₃ with α-Fe₂O₃ buffer layers are promising materials for α-Ga₂O₃ epitaxial growth.
ITO, which is a typical transparent conductive oxide, possesses two polymorphs; bcc-ITO (bixbyite) and rh-ITO(rhombohedral, corundum). Since the synthesis of rh-ITO requires high-pressure and high-temperature, few reports exist on rh-ITO thin films. Our group previously reported the epitaxial growth of rh-ITO using α-Fe₂O₃ buffer layers on various plane α-Al₂O₃ substrates [2]. However, detailed characterizations, such as dependence on the impurity concentration of rh-ITO thin film is not still investigated. Hence, we report on the characterization of the epitaxial rh-ITO thin films grown on various plane α-Al₂O₃ by mist CVD with α-Fe₂O₃ buffer layers.
The rh-ITO thin film prepared on the r-plane α-Al₂O₃ substrate with 5 at.% of Sn ratio in the starting solution possessed the resistivity of 1.73×10-4 Ωcm, the mobility of 25.8 cm²/Vs, and carrier density of 1.40×10²¹ cm^-3. Furthermore, the carrier concentration dependence of the mobility of rh-ITO epitaxial thin films was almost similar to that of single crystal bcc-ITO . In this symposium, we will discuss the detailed electrical and optical properties of rh-ITO on various α-Al₂O₃ substrates.

[1]N. Suzuki, K. Kaneko, S. Fujita, J. Cryst. Growth 364(2013), 30-33
[2]H.Nishinaka, M. Yoshimoto, Cryst. Growth Des. 18(2018), 5022-5028

Flexible electronic devices leading various wearable applications are expected to revolutionize modern society [1]. For the use of flexible functional materials, it has been mainly reported that the organic materials were formed on flexible substrates by low-temperature processes, such as inkjet printing [2]. However, organic materials generally exhibit low heat resistance and poor stability in a high-temperature environment. Conversely, flexible inorganic materials possess a high heat resistance and high stability in a harsh environment. However, it has been difficult to prepare single crystal thin films allowing to maximize the properties.
In this study, we demonstrated the epitaxial growth of oxide thin films on two-dimensional (2D) layered materials by atmospheric chemical vapor deposition (CVD) technique without requiring a vacuum process. The 2D layered materials can be simply peeled off and utilizes as flexible substrates. It has already been reported that oxide thin films were epitaxially grown on 2D layered materials by pulsed laser deposition (PLD) [3]. Moreover, the oxide epitaxially grown on the 2D layered materials maintained its characteristics even in a bent state or after repeated bending. We utilized the mist CVD methods for the epitaxial growth of oxide thin films on 2D layered materials. The mist CVD can be operated under atmospheric pressure for thin film growth because the stable mist including the CVD precursors is utilized instead of precursors vaporized by bubbling or heating. The mist is ultrasonically atomized from the solution made by dissolving the precursors. Furthermore, any solution precursor can be utilized as CVD precursors if it can be dissolved in solvents. Thus, one apparatus of mist CVD allows growing the various materials, as long as the solution precursors can be prepared.
We have demonstrated that the growth of metastable phases gallium oxide (Ga₂O₃) is controlled by mist CVD using other material buffer layer on 2D layered materials. Ga₂O₃ has recently attracted attention as an ultra-wide bandgap semiconductor in power devices and deep ultraviolet device applications. Ga₂O₃ possesses five polymorphs of α-, β-, γ-, δ- and ε-phase [4]. Ga₂O₃ thin films were grown on cleaved synthetic mica wafers as a 2D layered material substrate. A solution of gallium chloride (GaCl₃) in de-ionized water was used as the Ga precursor of mist CVD. X-ray diffraction (XRD) results revealed that ε-Ga₂O₃ thin films were grown on synthetic mica wafers at the growth temperature of 450-600°C without buffer layers. To obtain α-Ga₂O₃ thin films, we inserted α-Fe₂O₃ buffer layers grown by using iron acetylacetonate (Fe(C₅H₇O₂)₃) dissolved in de-ionized water as the Fe precursor [5]. As a result, the insertion of the buffer layers allowed growing epitaxially another metastable phase α-Ga₂O₃ thin films on synthetic mica wafers at the growth temperature of 350-600°C. We succeeded in epitaxial growth of the metastable ε phase and α phase Ga₂O₃ which are promising as power devices and deep ultraviolet devices on flexible substrates using mist CVD.
In the symposium, we will introduce and discuss the results of epitaxial growth on synthetic mica by mist CVD for not only several Ga₂O₃ phases but also various oxide thin films.
[1] Y. Liu, K. He, G. Chen, W. R. Leow and X. Chen, Chem. Rev., 117, 12893 (2017). [2] R. I. Haque, R. Vié, M. Germainy, L. Valbin, P. Benaben and X. Boddaert, Flex. Print. Electron., 1, 015001 (2016). [3] C. -I. Li, J. -C. Lin, H. -J. Liu, M. -W. Chu, H. -W. Chen, C. -H. Ma, C. -Y. Tsai, H. -W. Huang, H. -J. Lin, H. -L. Liu, P. -W. Chiu and Y. -H. Chu, Chem. Mater., 28, 3914 (2016). [4] R. Roy, V. G. Hill and E. F. Osborn, J. Am. Chem. Soc., 74, 719 (1952). [5] H. Nishinaka, D. Tahara, S. Morimoto and M. Yoshimoto, Mater. Lett., 205, 28 (2017).

Atomic layer deposition (ALD) is an emerging field in nanoscale materials research featuring atomic-level precision due to its self-limiting growth character. Besides fine thickness control, ALD also provides unmatched three-dimensional conformal and uniform film deposition. Owing to these advantages, ALD already has created a significant impact in aggressively scaled logic and memory device technology based on 3D CMOS transistors, as well as energy storage, solar cells, surface passivation, catalysis, and flexible/wearable devices.

ALD-grown films are vastly characterized via ex-situ measurements to quantify various material properties. However, gaining insight into the saturating surface reactions and growth mechanisms is only possible with real-time in-situ process monitoring of individual ALD cycles. While several in-situ measurement techniques have been employed in ALD research, in-situ ellipsometry stands out as one of the best options for real-time monitoring surface reactions. The promising potential of in-situ spectroscopic ellipsometry has already been demonstrated for a number of materials grown by remote plasma-ALD. Here, we verify that cost-effective multi-wavelength ellipsometer (MWE) can also be used effectively for real-time in-situ analysis of plasma-ALD growth cycles. We demonstrate for the first time that real-time dynamic in-situ MWE measurements convey not only accurate film deposition rate, but as well resolve single chemisorption and ligand exchange reactions with remarkable clarity. Moreover, forcing the limits for fitting the acquired in-situ MWE data, we were able to track the evolution of the optical constants of III-nitride films along the ALD cycles which showed thickness-dependent behavior.

Our main motivation behind this study is to analyze and compare the self-limiting growth characteristics of binary III-nitride (AlN, GaN, and InN) thin films via real-time in-situ ellipsometry and to gain insight into the ALD surface reaction mechanisms including chemical adsorption, ligand removal, and nitrogen incorporation steps.

Despite using the conventional alkyl metal precursors (trimethylaluminum, trimethylgallium, triethylgallium, and trimethylindium) utilized also widely in MOCVD epitaxial growth, their solid-gas surface interactions with nitrogen plasma species shows notable differences, particularly with respect to substrate temperature, plasma power, plasma exposure time, and plasma gas composition. In terms of substrate temperature, AlN exhibited crystallinity at lower temperatures when compared to GaN and InN. Even at 100 °C, AlN showed crystalline behavior whereas GaN displayed amorphous character up to 200 °C. While Ar/N₂/H₂ composition is optimal for AlN, N₂/H₂ and Ar/N₂ mixtures proved to be optimal for GaN and InN film quality, respectively. Furthermore, we have observed that the inclusion of H₂ gas into the plasma cycles led to substantial deformation of the hexagonal crystal structure and resulted in mixed phase growth with notable c-In₂O₃ and metallic tetragonal In. The possible surface reaction mechanisms that lead to these different growth behaviors will be discussed in detail.

The real-time in-situ data is correlated with film structural, optical, and chemical properties obtained by ex-situ spectroscopic ellipsometer, x-ray diffraction (XRD), x-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), and transmission electron microscopy (TEM) measurements. Our results demonstrate that dynamic in-situ MWE measurements can provide significant insight about the surface reactions taking place during low-temperature plasma-ALD of III-nitride compounds.

Zr doped-hafnium oxides (HZO)-based FeFET/FRAM has recently emerged as the next-generation memory device due to its low power operation, scalability, and Si compatibility. However, it suffers from the limited processing window to secure their ferroelectric properties in terms of thickness, thermal processing effect and high dependence on ozone (O₃) as the oxygen precursor of atomic layer deposition (ALD). Herein, we report on the use of hydrogen peroxide (H₂O₂) precursor as the O₂ source to fabricate HZO thin films via ALD. We optimized the deposition parameters such as purging time, deposition time, and temperature to obtain high and stable ferroelectric properties. Furthermore, we evaluated the thermodynamic reaction energy of hydrogen peroxide and compared the two classes of samples in terms of deposition state, thickness, and chemical composition using X-ray photoelectron spectroscopy (XPS). To analyze the ferroelectric properties of HZO thin films, we characterized the crystalline structure using X-ray diffraction (XRD), and the electrical properties using C-V and P-V measurements. Finally, we investigated the grain size distributions using atomic force microscopy (AFM) and correlated them with their respective crystalline phase. Through this study, we provide a new ALD technique for high-quality HZO thin films for FeFET/FRAM device applications.

Recent results are presented from studies of the growth by Chemical Vapor Deposition of Carbon Nanotubes (CNTs), using a novel technique which it is hoped will ultimately lead to the availability of ultra-long CNTs. The method used involves incorporation of refractory, high-melting-point metals into the nano-particles of catalyst metal from which the CNTs nucleate and grow. The refractory metal acts as a diffusion inhibitor, and inclusion of these diffusion inhibitors slows the erosion of the catalyst particles, and thus extends the lifetime of the catalyst particles and hence the length to which CNTs can be grown. This effect has previously been demonstrated for several catalyst/diffusion-inhibitor systems including Mo/Re and Fe/Mo. Here we present new results on the Fe/Re catalyst system: it is shown that inclusion of Re diffusion inhibitor can enhance the lifetime of the iron catalyst particles and give CNT growth to lengths greater than would otherwise result.

Conformal polymer thin films have a wide variety of uses ranging from protective coating applications to surface functionalization and modification. There are a variety of traditional wet methods for depositing polymer films, such as spin coating or drop casting, but substrates with complex features will have issues caused by varying wettability and capillary forces. Initiated chemical vapor deposition (iCVD) is a low-temperature, dry process that deposits conformal polymer thin films from the gas phase. The process uses an initiator compound that is activated by a hot filament array and polymerizes adsorbed organosilicon precursor on the substrate surface. For this study the reactor parameters have been optimized to maximize conformality of the coating at multiple length scales.

Coating features with irregularities at multiple length scales is a challenge that is commercially relevant both for scaling the process and improving reliability of the coating across a diverse range of substrates. To test conformality and study the effect of geometry on the deposition process high-aspect ratio (>10:1) millimeter and micron scale testing devices were fabricated on silicon. In both length scales, lowering the sticking probability improved conformality of the coating, which could be achieved by varying the temperature of the substrate and reactor pressure. On the millimeter scale, it was clear that the conformality was reaction limited by depletion of the initiator, whereas film growth in the micron-scale features was limited by monomer diffusion and adsorption. Conditions for optimal conformality at both length scales was demonstrated with different sets of reactor conditions which can be combined into a multi-step conformal coating process.

Graphene oxide (GO) membranes have shown promising ionic and molecular sieving due to their unique interconnected nanochannels;
however, their low salt rejection efficiency caused by an enlarged interlayer spacing between GO flakes limits their wide application in
water desalination. For the purpose of enhancing GO membrane performance, this research applies a novel approach to atomic level surface functionalization with precise control, which allowed for the formation of an ultra-thin Al2O3 layer by sequential atomic layer deposition (ALD) to the GO membrane. Due to the strong chemisorption of trimethylaluminum (TMA) to GO, an ultra-thin (1.44 nm) and uniform hydrophilic Al2O3 layer was successfully deposited on the GO membrane under low numbers of ALD cycles. The GO membrane coated with an atomically thin and uniform Al2O3 layer achieved enhanced water permeability (from 40 LMH/bar to 70
LMH/bar) and NaCl rejection up to 67.4 %, successfully overcoming the trade-off between permeability and rejection efficiency. This enhanced water permeability is mainly attributed to the increase in surface hydrophilicity without narrowing the GO membrane pore structure. Dspacing via X-ray diffraction (XRD), surface chemistry via Fouriertransform infrared spectroscopy (FTIR), and water contact angle (WCA) measurements confirmed the effect of ALD on the physico-chemical properties of the GO membrane. The distinctive features of ALD to the GO membrane may contribute towards enhancing its wider application in water desalination.

This talk will discuss our use of vapor phase infiltration (VPI) to alter the optoelectronic properties of poly(aniline) (Pani). VPI is similar to atomic layer deposition (ALD), but instead of cycling viscous flows of co-reactants, VPI doses a single reactant vapor in a static atmosphere followed by a purge-pump cycle. ALD typically provides conformal surface coatings while VPI enables extensive sub-surface diffusion and reactions within organic substrates. In our study, Pani was infiltrated with a single cycle of TiCl₄-H2O to yield a hybrid film. This chemistry has not been explored before and was specifically selected because of its reaction products; TiO_x has useful optical properties for photovoltaic and photochemical applications while HCl is known to dope Pani and create polaronic charge carriers. We found that Pani films optimally infiltrated with TiCl₄ have an electrical conductivity of ca. 0.2 S/cm. To the best of our knowledge, this is the highest electrically conducting Pani film achieved with a single-cycle exposure to a metal-halide vapor. Additionally, as VPI hold time increases, Pani films turn from blue to nearly transparent to green. This color transition is believed to be evidence of the generation of polaronic charge carriers and the concomitant formation of TiO_x. The polaronic carriers absorb the red-NIR-MIR spectra (700- 5000 nm) while TiO_x contributes to unusually high absorbance in the blue-UV region (450 - 350 nm). This combination of optical absorbances in Pani has not been documented before and could be valuable for electrochromic or photovoltaic applications. To better understand processing parameters, we varied the TiCl₄ exposure time and measured ex-situ optical signatures. We calculated an effective TiCl₄ diffusion coefficient of ca. 10^-15 cm²/s at 80° C. Additionally, we demonstrated that the relative ratio of UV and IR absorbing species can be controlled by varying reaction temperature. Increasing the reaction temperature likely decreases HCl sorption, ergo decreasing the extent of doping, polaron formation, and IR absorbance. Lastly, we performed XPS measurements to elucidate a reaction mechanism. We found that titanium effectively only binds with oxygen and that nitrogen radical cations (polarons) are present; this further corroborates our hypothesis that TiO_x is responsible for the unique blue-UV absorbance and that the electrical conductivity is dominated by polaronic carriers. By better understanding this reaction mechanism and kinetics, VPI precursor chemistry and processing conditions can be optimized for future organic semiconductor doping.

Nanocellulosic films (nanopapers) are of interest for packaging, printing, chemical diagnostics, flexible electronics and separation membranes. These nanopaper products often require chemical modification to enhance functionality. Most chemical modification is achieved via wet chemistry methods that can be tedious and energy intensive due to post-processing drying. Here, we discuss the use of atomic layer deposition (ALD), a vapor phase modification technique, to quickly and simply enhance the wet strength and durability of nanopaper. Here, we specifically consider ALD processes that use only a “few” cycles (< 10) to maintain the potential for process scalability and commodity manufacturing. While these “few cycle” ALD treatments do not appear to alter the mechanical properties of the nanopaper in the dry state, wet strength shows a significant enhancement. Here we will report that up to 10 cycles of either aluminum oxide or titanium oxide ALD is sufficient to significantly increase the durability of nanopaper under aggressive aqueous sonication or burst testing (9x increase). In this presentation, we will discuss our investigation into whether this increase in wet strength is the result of enhanced hydrophobic attractions or increase hydrogen bonding attraction. Through a combination of experimental investigations and ab initiocalculations, we currently believe the M-OH terminations enhance hydrogen bond strength between the neighboring cellulosic nanofibrils.

Atomic layer deposition (ALD) is a highly useful synthesis method to grow highly-controllable thin-film materials. In a typical ALD process, the initial nucleation and growth is of particular importance, because it not only is crucial for preparing high-quality film–substrate interfaces but also can be engineered to synthesize well-controlled nanoparticles. Therefore, understanding the initial surface chemistry as well as the growth mechanism is of crucial importance for rationally designing functional materials and interfaces by ALD. Over the past several years, ALD of metal sulfides has aroused great attention, especially for their numerous applications in the emerging energy technologies. Many new ALD processes of metal sulfides have been recently developed, and the metal sulfides of which the ALD processes were only recently available include GaS (2014), GeS (2014), MoS₂ (2014), Li₂S (2014), Co₉S₈ (2015), NiS (2016), MnS (2016), FeS (2017), VS₄ (2017), Bi₂S₃ (2017), AlS_x (2017), and ReS₂ (2018). The investigation of the initial growth on these new metal sulfide ALD processes is certainly valuable and in need.

Our group recently developed a new ALD process of nickel sulfide (NiS) using bis(N,N’-di-tert-butylacetamidinato)nickel(II) (Ni(amd)₂) as the nickel precursor and H₂S gas as the sulfur source. ^[1][2] Herein, we study the initial ALD growth of NiS on SiO_x, by using the in situ techniques of X-ray photoelectron spectroscopy (XPS) and low-energy ion scattering (LEIS).^[3] The reaction scheme during the initial ALD stage is found to be rather different from that in the later steady film growth stage, and therefore a reaction-agglomeration mechanistic scheme is proposed to describe the initial ALD growth. In the scheme, the conversion from Ni–O to NiS in the H₂S half-cycles is suggested to be accompanied by the spontaneous agglomeration of the ligand-stripped Ni-containing species to afford NiS clusters, and the agglomeration can re-expose the surface of SiO_x and therefore allow the surface to further react with Ni(amd)₂ in the subsequent ALD cycles. Using LEIS, the transition from the initial reaction-agglomeration growth to the continuous steady film growth is found to be between 100 and 300 ALD cycles. Ex situ atomic force microscopy (AFM) and transmission electron microscopy (TEM) are further employed to corroborate the presence of the agglomeration during the initial growth. The results reported herein and the implied growth mechanism should be of great representativeness for the ALD of metal sulfides, and therefore, the findings should be highly valuable for future engineering of functional metal sulfide nanomaterials and interfaces by ALD.

References
[1] H. Li, Y. Shao, Y. Su, Y. Gao, X. Wang*, Chem. Mater. 28, 1155 (2016).
[2] R. Zhao, Z. Guo, X. Wang*, J. Phys. Chem. C 122, 21514 (2018).
[3] R. Zhao, X. Wang*, Chem. Mater. 31, 445 (2019).

Development of wide band-gap p-type oxide semiconductors with good properties is essential for various applications that realize resource saving and low power loss. Although oxide semiconductors that show p-type conductivity with wide band-gap are rare, nickel oxide (3.7 eV) has attracted attention. In recent years, p-type oxide semiconductor materials having a larger band gap are required. Therefore, we focused on Ni_1-XMg_XO, which is a mixed crystal with MgO, which has the same rock salt structure as NiO and has a band gap of 7.8 eV. In addition, in order to improve the properties of these materials, growth was performed using a mist CVD method that enables high-quality film growth even under atmospheric pressure.
In this presentation, Ni_1-XMg_XO thin films grown by using the mist CVD method will be described in detail. Ni_1-XMg_XO epitaxial growth was performed on α-Al₂O₃ or MgO substrates, and carrier concentration control by Li doping was studied. The bandgap of Ni_1-xMg_XO monotonously increased according to the Mg composition and was controlled in the range of 3.4–4.0 [eV]. It was revealed that the carrier concentration in the Li-doped NiO thin film can be controlled in a wide range only by controlling the Li concentration in the precursor solution.
A wide band-gap p-type oxide semiconductor Ni_1-XMg_XO was grown by the mist CVD method. These are very high quality despite being grown in a non-vacuum process and are expected to be applied to various devices.

Layered transition metal dichalcogenides (TMDCs) such as a layered MoS₂ are attracting great attentions owing to their outstanding physical, chemical, and mechanical properties, compared with their bulk counterparts. Chemical vapor deposition (CVD) is known to be the most effective way to synthesize the layered MoS₂ on the target substrate. While previous studies suggested to use different types of reactants/precursors like sulfur powders, and gaseous H₂S and H₂ precursors for a higher-quality MoS₂ layer, their reaction kinetics and mechanisms have yet to be fully understood. Here, we perform reactive molecular dynamics (RMD) simulations for CVD synthesis of MoS₂ layer using three different systems: 1. MoO₃ and S₂; 2. MoO₃ and H₂S; 3. MoO₃ and H₂S/H₂ mixture. Our goal is to clarify atomic-level reaction pathways for reduction/sulfidation of the MoO₃ slab using different gaseous environments. RMD results reveal that MoO₃ slab could be effectively sulfurized by the H₂S/H₂ mixture. As such, our computational work provides a valuable input for experimental synthesis of TMDCs using the CVD technique.

This work was supported as part of the Computational Materials Sciences Program funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award Number DE-SC0014607. The simulations were performed at the Argonne Leadership Computing Facility under the DOE INCITE and Aurora Early Science programs, and at the Center for High Performance Computing of the University of Southern California

Oxidative Chemical Vapor Deposition (oCVD) and Atomic Layer Deposition (ALD) accomplish polymers and oxides synthesis, thickness-defined patterned deposition, and doping in a single process at near-room temperatures (40 - 80 °C) for a broad range of technological applications. Also, both function independent of material solubility and substrate properties to enable insoluble polymers and oxides to be coated in a conformal or uniform manner without any solvent damage to the substrate. This key differentiator from solution-based approaches makes them substrate-independent scalable platform technologies for surface modification and device fabrication.
Based on the commercialization-critical benefits, semiconducting and conducting polymer thin films with suitable properties, patterns, and size can be directly integrated into flexible electronics such as organic photovoltaics on unmodified papers, thereby enhancing the efficiency and lifetime significantly. Meanwhile, H⁺ conducting and O₂ impermeable ultrathin (2 nm) SiO₂ membrane leads to runaway H⁺ flux (3000 nm^-2 s^-1) faster than maximum solar photon flux (1500 nm^-2 s^-1). As a result, the SiO₂ membrane can act as an enhanced charge-carrier platform, where pushing artificial photosynthesis beyond mass (H⁺) transfer limits, and thus making it scalable to the unprecedented level of several terawatts.

Solid electrolytes are one of the key parts in the field of modern energy technology. While zirconia based oxides are used as electrolytes for e.g. solid oxide fuel cells (SOFC) or oxygen sensors, ceria-based solid solutions are promising for applications for electrode materials in SOFCs and heterogeneous catalysts. Great efforts are made to enhance the electrical and / or the ionic conductivity of these solid electrolytes. The transport properties of these oxides can be altered by varying the composition or by nano- and microstructuring the materials. For the later, the primary reason is the increased density of interfaces and/or surfaces. In recent years, enhanced conductivity in nanostructured oxides compared to their bulk counterparts has been observed, but the underlying mechanisms are still not fully understood. This work aims at evaluating the surface effects on the transport properties of nanostructured thin film oxide ceramics. Therefore, a suitable route to produce single-phase mesoporous thin films and two-phase-nanocomposites thereof has to be established.

In this work, we present a method to prepare two-phase nanocomposites made of different types of oxides. The method of Atomic Layer Deposition (ALD) was used to deposit a thin layer of different oxides, e.g. Al₂O₃ or TiO₂, into a mesoporous thin film of yttria-stabilized zirconium oxide (YSZ). Thin porous films were synthesized by an evaporation-induced self-assembly (EISA) process using the dip coating method. Following this route, a mesoporous film of 8mol% YSZ was produced with a pore size diameter of 18 nm and a well-ordered pore structure. This porous film was used as the host material for the Atomic Layer Deposition. The process of ALD is so far the best technique to coat a three-dimensional structure having a high-aspect ratio with a thin, homogeneous and conformal layer along with highly precise thickness control. Up to date, there are only few studies about an ALD process into ordered mesoporous structures. In this context, a new ALD process is established to coat the three-dimensional surface structure of a highly ordered mesoporous thin film. Processing parameters were optimized to enhance the diffusion of the precursor molecules through the entire mesoporous network. The characterization of the coated mesoporous film was done by both scanning electron microscopy (SEM) and transmission electron microscopy (TEM). X-ray reflectometry (XRR) was used to measure the growth rates of this specific ALD process. To ensure a complete coating on the surface throughout the entire mesoporous film, a cross-section of the film was measured by atom-probe tomography (ATP). Finally, the effect of the surface coating on the transport properties were investigated by impedance spectroscopy. The modification of the pore surface structure was found to affect the conductivity of the mesoporous thin film.

Zinc Oxide (ZnO) has been shown to exhibit semiconducting and piezoelectric dual properties. This has led to a large commercial demand of ZnO for optoelectronics that operate at the blue-ultraviolet regions. Consequently, varying techniques have been devised to create different nanostructures of ZnO. Here, the single step synthesis of ZnO nanostructures was performed on Si(100) substrates with a thin ZnO seed-layer. A modified chemical vapor deposition (CVD) method was developed to accomplish the structure formation. Sb doping of the structures in the gas phase was performed to study its effects on structure and optoelectronic properties. Different structures were realized including nanofilaments, nanoparticles, microflowers, nanorods, and nanocolumns. Only nanorods/columns are shown in this work. Morphology was examined using scanning electron microscopy (SEM). Energy-dispersive X-ray spectroscopy (EDS) and X-ray powder diffraction (XRD) were used for structural studies. Optoelecronic properties were explored using room-temperature photoluminescence (PL) spectroscopy. PL data show the relative increase in the number of defects and decline in crystal quality upon dopant introduction. Some structural defects might be attributed to the diffusion of Sb ions into the lattices of ZnO, replacement of Zn by Sb, and ionic radii difference. These stacking faults are most likely the reason behind the dominance and broadening of DLE peak.

Metal Ir films have been widely used as electrodes, especially for oxide capacitor including ferroelectric random access memory. For these applications, chemical vapor deposition (CVD) is a most promising process because three dimensional structure capacitor structures are essential to increase the capacitor area on the limited foot print of the substrate. We developed spray chemical vapor deposition, in which CVD source was spray on the heated substrate in air [1-4]. By using this process, three dimensional films of indium tin oxide and Ru with good step coverage were successfully obtained on the trenched substrates. In the present study, we apply this spray CVD to Ir metal films deposition using novel of Ir precursor, (1,3-cyclohexadiene)(ethylcyclopentadienyl)iridium, Ir(EtCp)(CHD).
Ir(EtCp)(CHD) is a liquid source at room temperature and exhibited enough vapor pressure (0.1 Torr/75^oC), excellent volatility and adequate decomposition temperature. It is already ascertained that Ir metal films grown using Ir(EtCp)(CHD) showed shorter incubation time and higher nucleation density than those of films using Ir(EtCp)(CHD)[5]. Ir(EtCp)(CHD) was dissolved in ethanol and the concentration was fixed at approximately 0.05-0.1 mol/l. The solution was sprayed manually on the substrates in air with an inexpensive atomizer used for cosmetic purposes. The substrates were heated on a hot plate at 200–430 ^oC. We used non-flat Si substrates (patterning SiO₂/Si, opening depth: 1 μm, opening sizes: less than 1 μm) and plane SiO₂/Si substrates.
Film deposition was ascertained on SiO₂/Si substrate at 240-430 ^oC and deposition rate increased with the deposition temperature, but was saturated above 330 ^oC. Obtained films consist of Ir metal without any oxide impurity irrespective of the deposition temperature. Films tend to orient to (111) with increasing deposition temperature. Resistivity of these Ir metal films was on the order of 10^-6 Ωcm, which was on the same order with that of bulk Ir metal. Good step coverage was observed for Ir metal films deposited at 270 and 330 ^oC. This shows that simple spray CVD process deposited in air is a good way to deposit Ir metal films with good conductivity and step coverage.
[1] Y. Sawada et al., Thin Solid Films, 409 (1) (2002) 46-50.
[2] T. Kondo et al., J. Crystal Growth 311 (2009) 642–646
[3] T. Kondo et al., , Electrochem. Solid-State Lett., 12 (2009) D42-D44.
[4] T. Kondo et al., Thin Solid Films 516 (2008) 5864–5867
[5] K. Kawano, et al, Mat. Rec. Soc. Symp. Proc., 784 (2004) C3.30.1-6

Vertically aligned graphene nanosheet arrays (VAGNAs) exhibit large surface area, excellent electron transport properties, outstanding mechanical strength, high chemical and thermal stability, and enhanced electrochemical activity, which makes them highly promising for application in supercapacitors, batteries, fuel cell catalysts. It is shown that VAGNAs terminated with a high-quality single-layer graphene sheet, can be directly grown on germanium by plasma enhanced chemical vapor deposition without an additional catalyst and at low temperature, which is confirmed by high-resolution transmission electron microscopy and large-scale Raman mapping. The uniform, centimeter-scale VAGNAs can be used as a surface-enhanced Raman spectroscopy substrate providing evidence of enhanced sensitivity for rhodamine detection down to 1×10-6 mol L-1 due to the existed abundant edges.

Single-walled carbon nanotubes (SWNTs) have been expected to be applied to various fields such as nanoelectronic devices, transparent conducting films, energy devices and sensors due to their outstanding physical and chemical properties. A chemical vapor deposition (CVD) is the most popular method for SWNTs synthesis because of its simplicity on SWNTs synthesis process and easy control of process parameters. In general, most of the synthesis experiments have been performed in the central region of the reactor so far where temperature uniformity is guaranteed.
In this study, we report the result of detail investigation of the SWNTs synthesis yield depending on sample position in a horizontal CVD system. Methane and Fe thin films were used as feedstock and catalyst for SWNTs synthesis, respectively. Ultra high resolution scanning electron microscope (UHR-SEM) was used to confirm the synthesis yield variation of SWNTs along the axial distance of the reactor. The morphology and crystallinity of the synthesized SWNTs were evaluated by atomic force microscope and Raman spectroscopy, respectively. Main result of this study shows that the highest synthesis yield of SWNTs was observed at the rare region of the reactor not the central region. The results of this study are expected to be applicable to the synthesis of various nanomaterials using CVD process.

Single-walled carbon nanotubes (SWNTs) have attracted much attention due to both their outstanding physical properties and their nanoscale dimensions. Although SWNTs can be obtained by several processes, the most prevailing method to control tube structure, growth direction, diameter, and chirality, would be a thermal chemical vapor deposition (TCVD). However, the high temperature operation of this method limits the industrial accessibility. Plasma-enhanced chemical vapor deposition (PECVD) is a well-known for growth temperature reduction although the SWNTs grown by PECVD are generally perpendicular to the substrate in conventional vertical chamber along the electric field line, which limits the further manipulation of SWNTs.
We here demonstrate the low temperature growth of SWNTs using a plasma-assisted TCVD system. The system is made of 1 inch horizontal quartz tube and composed of inductively coupled plasma (ICP) production region for efficient decomposition of feedstock and TCVD region for SWNTs growth region. The growth substrate installed at the center of the TCVD region and varied the distance from the plasma generation region to find avoid probable ion damages. We used iron thin film and ethylene gas as growth catalyst and feedstock, respectively. Optical emission spectroscopy was used to analyze ion and active species in the plasma with respect to the process parameters, such as plasma power, pressure, gas composition, and distance from the plasma formation region. The detail of the results will be discussed at the meeting.

Bulk 2D layered materials are promising material platforms to study their properties, surfaces, chemical reactivity, as well as potential applicability in diverse applicative areas. Some of the most commonly researched 2D layered materials under this category are: graphene, boron nitride, black phosphorous, and transition metal chalcogenides (TMCs).
Among the TMCs, silver selenide is of particular interest because this promising A^I₂B^VIsemiconductor can be used to make solid-state electrochemical sensors, solar cells, magnetic field sensors, optical filters, etc. [1]. Some of the established synthesis methods include the use of micro-emulsion mixing, microwave radiation, sonochemistry, and electrochemical potential to drive the decomposition of selenium and/ or silver compounds to form Ag₂Se [2].
Although all these synthetic approaches are significant in their respective fields, in our case, we have synthesized Ag₂Se using an atmospheric pressure chemical vapor deposition (AP-CVD). We grew the material on a silver foil in an inert atmosphere, using elemental selenium as the precursor. We characterized the final product using XRD, HR-SEM, TEM, and Elemental Mapping (EDX) to ascertain its stoichiometry and we concluded that we obtained β-Ag₂Se (low-temperature phase). To our knowledge, this is the first report of β-Ag₂Se grown using AP-CVD with this specific morphology to be tested as a prospective candidate for oxygen reduction reaction (ORR). Since any non-platinum catalyst is always considered a viable and cheaper alternative, our idea behind the preparation of this material was to test it for ORR and to find out its applicability towards fuel cells. From the linear sweep voltammograms (LSV), we observed that β-Ag₂Se exhibited -0.2 mA current in 1 M KOH solution at 400 RPM, which is quite higher than that observed for 20 wt.% Pt/C, i.e., -0.2 mA. Again with increasing rotating speed of the rotating disc electrode (RDE), we saw that the current density for both the studied catalysts increased due to the effect of higher mass transfer to the working electrode. Furthermore, our studied catalyst, i.e., β-Ag₂Se exhibited -0.55 mA current in 1 M KOH solution at 2400 rotations per minute (RPM), whereas commercial Pt/C shows -0.2 mA current under the same conditions. These encouraging results point to the unique structural 2D layered morphology of β-Ag₂Se to be responsible for the improved electrocatalytic activity towards ORR in the alkaline medium compared to that of the state-of-the-art cathode catalyst Pt/C.

References:
[1] Li et.al/Journal of Physical Chemistry C, 112, 8, (2008), 2845-2850
[2] Cui et.al/Journal of Power Sources, 327, (2016), 432-437

Aligned multiwall carbon nanotube (CNT) arrays on conductive substrates have been envisaged as interconnects in microelectronics and ultra-high density magnetic information storage material in spintronics. Such applications require the CNTs to be grown vertically aligned, and in direct contact, to conductive substrates [1]. This could be achieved by controlling the physical aspects related to the dewetting of the metal catalyst leading to the formation of a stable assembly of self-organized nanoparticles suitable to catalyse CNT growth. In addition, the use of plasma enhanced chemical vapour deposition (PECVD) allows controlling the gas phase chemistry. In these directions, we propose a simple and reliable fabrication approach to produce controlled cobalt (Co) magnetic nanoparticles by solid state dewetting of ultra-thin Co films deposited on a conductive titanium silicon nitride (TiSiN). TiSiN layers act as barriers to limit diffusion into the bulk of the substrate. Moreover, the low surface energy of TiSiN favours catalyst dewetting, thereby improving CNT forest density and verticality. To control gas phase chemistry, (CH₄/H₂/O₂) mixture gases were decomposed over the catalyst nanoparticles after their dewetting. The addition of oxygen leads to the formation of water H₂O and OH radicals that improves CNT quality.

Experimentally, we achieved the dewetting of cobalt films with thicknesses h= 1, 2 and 3 nm by thermal heating at 750°C followed by hydrogen plasma treatment. This resulted in assembly of nanoparticles with quasi spherical shapes and well-controlled size distribution. The mean particle size (D) have been measured from scanning electron microscopy and correlated with the initial film thickness, namely D [endif]--> 6h. Moreover, using the TiSiN barrier allowed obtaining homogeneous nanoparticle distribution with density as high as ~ (4.2 ± 0.1) ×10¹¹ cobalt nanoparticles/cm². This value is smaller than the one reported for iron ~ (5.1 ± 0.1) ×10¹² [1]. However, cobalt is more stable regarding oxidation and thus presents a higher magnetization at saturation. After addition of (H₂:CH₄:O₂) (90:10:x) sccm at 10 mbar, PECVD allowed to obtain vertically aligned multi-walled nanotubes. Oxygen flow rate x was varied from 0 to 4 sccm and correlated to the CNT growth rate and surface density. At 2 sccm of oxygen, ultra-dense ([endif]-->10¹¹ nanotubes/cm²) vertically aligned CNTs with well graphitized walls as characterized by high resolution transmission electron microscopy (HRTEM) were obtained. Tow numerical models were proposed to understand the experimental results. As a first approximation, the plasma is described by spatially averaged bulk properties in (0D). This model extends the classical chemistry formulation to nonequilibrium plasma reactors for CNT growth from 134 species in (C,H,O,e-) system involved in 471 gas-phase reactions [2]. The small computational demands of (0D) model, allows kinetics sensitivity analysis and permit the identification of dominant reactions and key species depicting CNT growth. The reduced chemical scheme was then used within (2D) model including 22 gas species and 100 homogeneous reactions. Both gas and surface chemistries were taken in consideration to calculate CNT growth rate using Surface-Chemkin and Ansys-Fluent softwares. Surface-reaction mechanism specific to cobalt under (H₂/CH₄/O₂) CVD plasma is proposed and considers 16 surface species and 81 surface reactions involving both gaseous species impinging on the surface, adsorbed molecules, atoms and free sites. The bulk material considered is CNT, with mass density of 2.2 g/cm³ and site density ~3.6 ×10^-10 moles/cm² as estimated from measured nucleation surface density, CNT diameter and number of walls. Optimal CNT rate was found for 2 sccm O₂ in good agreement with experimental measurement.


1. J. Yang et al., Applied Physics Letters, 2015. 106(8): p. 083108.
2. K. Pashova et al., Plasma Sources Sci. Technol., 2019. 28: 045001

Chemical vapor deposition (CVD) is the most promising approach to obtain high-quality, large-area graphene. Recently, bilayer graphene has attracted massive interests owing to its promising applications as twistronics, superconductors, etc. To obtain graphene layers beyond monolayer, counter-intuitive asymmetric CVD growth is developed to break the original self-limiting nature of growth on copper substrate (Nat Nano 11, 426 (2016)). However, the layer control in graphene remains challenging and lack of study. Researchers fail to obtain pure, high-quality large-scale bilayer graphene and have thus perceived that thicker graphene impurities are inevitable, after more than one decade of research efforts.
Here, we report a technique to control the graphene layer number in asymmetric CVD growth. Owing to counter-intuitive growth principle, we proposed a new physical quantity, named “interface adhesive energy”, that can be used to predict the CVD growth behavior. We show that, through first-principle calculations, this new physical quantity can used to control the resulting graphene layer number. Based on these theoretical understandings, we have devised a new CVD strategy, which results in layer control in asymmetric CVD bilayer graphene growth. Through systematic materials characterization studies (SEM, XPS, Raman, STEM, and AFM), we found that the growth of CVD bilayer graphene on copper catalyst consists of several major steps: (i) dehydrogenation of carbon precursor (i.e.CH₄) at the Cu interior resulting in free carbon atoms, (ii) bulk diffusion of carbon atoms from inner to outer surface of the Cu pocket, (iii) interface diffusion of carbon atoms between the 1^stLG and Cu exterior, and (iv) nucleation of graphene adlayers (i.e.2^ndLG, 3^rdLG, 4^thLG etc.) or growth continuation of the existing graphene layer (i.e.2^ndLG). Through Raman studies, we confirmed that our bilayer graphene is of AB stacking with small twisting angles (θ = 0 - 5°), namely quasi AB-stacked bilayer graphene. Specifically, our samples exhibit 4 Raman characteristic signatures: the positions of G bands ~1580/cm, the positions of 2D bands ~2680/cm, 2D/G intensity ratio ~1, and 2D bands’ full-width-half-maximum ~58/cm. We also proposed an unsupervised learning approach to manage large-volume Raman data. Our artificial-intelligence-assisted Raman analysis approach is very useful for graphene samples in identifying regions of different stacking orders and layer numbers.
In summary, by introducing the concept of interface adhesive energy, we managed to prevent the growths of thicker graphene islands of impurities and realized the long-sought layer control in graphene CVD growth. Confirmed by artificial-intelligence-assisted Raman analysis, the obtained graphene layers were in AB stacking with small twisting angles. Our discoveries on well-controlled growth and coherent characterization of large-area graphene are essential for unlimited innovations based on two-dimensional materials, owing to their electrical, optical, mechanical and chemical properties.

Nowadays device applications of amorphous silicon carbide (a-SiC:H) include light emitting diodes, phototransistors, photodetectors, MEMS and solar collectors. The a-SiC:H films have different optical, electrical and structural characteristics depending on the gases mixtures and the deposition technique used. The plasma enhanced chemical vapor deposition (PECVD) technique has several advantages, as its low temperature of deposition (< 200°C), which implies the use of different substrates, as plastics and metal foils, as well this technique produce uniform films.
However also it has been reported that as deposited a-SiC:H films have instability and this could affect the photoluminescent (PL) emission of a-SiC:H films. One way to reduce the instability and defects is by thermal treatments. In conventional thermal treatments the films are exposed to high temperatures for long periods of time in an inert environment, while in the rapid thermal annealing (RTA) the substrates are subjected to very short periods of time (seconds).
In this work we present the effect of thermal processes and rapid thermal annealing on the PL emission of amorphous silicon carbide (a-SiC:H). The a-SiC:H thin films were deposited at a substrate temperature of 150 °C by PECVD, silane (SiH4) and methane (CH4) gases were used. The rapid thermal annealing (RTA) was carried out with a ramp up of 50°C/s at 500 °C for 40 seconds and with a ramp down 120 seconds. The conventional thermal treatment was done with a temperature of 1100 ° C for 5400 seconds in forming gas environment.
Our results shows that the a-SiC:H films as deposited, present emission in orange region of the visible spectrum, while the films treated with RTP have an increase in the PL emission intensity, with an emission in the visible region (in the blue color), that increment is attributed to radiative centers related to oxygen. On the other hand with conventional thermal treatment processes the emission is decreased with respect to the deposited films. The above is related to the structural change suffered by the films when are exposed to thermal treatment process. Finally, our results show that RTA is more efficient than conventional thermal treatments in order to increase the PL emission of a-SiC:H films, which is of interest for light emission applications.

Flexible and transparent electronics have presented a groundbreaking era for the world of display technologies. Zinc oxide (ZnO) amongst the popular transparent conducting oxides are the most commonly used semiconductor for transparent devices, such as liquid crystal displays, solar cells, transparent heaters, and etc. Liquid crystal displays most significant component are thin film transistors (TFT). Modern day TFTs are produced with an active layer of either hydrogenated amorphous silicon(a-Si:H) or indium gallium doped zinc oxide (IGZO). A-Si:H will soon be out shadowed by doped ZnO counterparts due to its minimal channel mobility. IGZO contains indium, which is a toxic, expensive, and whose supply will not be enough for demand by 2020. The indium free oxide-based channel material such as titanium doped zinc oxide (TiZnO) has been immensely studied due to its good electrical and optical properties. High quality TiZnO thins films were grown by atomic layer deposition on n-type silicon, glass, and sapphire substrates. The films were varied with different dopant ratios to properly evaluate and analyze the effect of Ti substitutional atoms in the crystal lattice. The effect of Ti content on microstructures, surface morphology, electrical properties, and optical properties of the films were investigated by Atomic Force Microscopy (AFM), X-Ray Diffraction, Ultra-violet Visible Spectroscopy (UV-VIS), and Hall Effect Measurement System (HMS). AFM provided thin films roughness, grain size, and surface morphology for glass, silicon, and sapphire thin films. XRD provided the effect of how different Ti substitutional atoms doping ratios invaded the crystal lattice. UV-VIS was performed to investigate the high optical transparency for the glass substrates. HMS provided the electrical mobility, conductivity, and resistivity for the active layer of TFT TiZnO thin films. The present work will provide valuable scientific input for TiZnO thin films for the advancement of TFT devices.


This work was supported by the NSF-CREST Grant number HRD 1547771 NSF-CREST Grant number HRD 1036494

Atmospheric pressure spatial atomic layer deposition (AP-SALD) is a fast and scalable thin film deposition technique that operates in atmospheric pressure. It has the ability to control the film thickness in the atomic scale range and produce uniform and pinhole-free films at modest temperatures (e.g. 80-200°C). AP-SALD is implemented to deposit an insulator film with a thickness gradient as opposed to a film with uniform thickness. To obtain a thickness gradient of metal oxide layers such as Al₂O₃ or ZnO on Si or glass substrates, we adjusted the distance between the deposition reactor head and the substrate to be 100µm on one side and 150, 190, or 230 µm on the other side. The head-substrate separations used here are sufficiently large that some mixing of the precursors occurs in the gas-phase, resulting in some chemical vapour deposition (CVD). By varying the head-substrate separation across the substrate, the amount of CVD was varied, resulting in different deposition rates across the substrate. To demonstrate the applicability of this technique, combinatorial MIM diodes are fabricated, i.e. diodes with many different thicknesses can be fabricated on a single substrate. The diode performance showed variation with aluminum oxide thickness and the optimal thickness is identified.

Pressure sensitive adhesives (PSA) are materials that can bind different surfaces together. They can easily adhere to various surfaces at a light pressure at room temperature. Common acrylate monomers which are used to produce PSA materials are ethylhexyl acrylate (EHA), methyl methacrylate (MMA), and acrylic acid (AA). PSAs based on acrylate polymers are widely used because of their useful properties such as high bonding strength, easiness of application to various surfaces, and low manufacturing cost. Commercially, PSAs are deposited using wet techniques. In this study, P(EHA-co-AA) copolymer thin films were deposited on different surfaces using initiated chemical vapor deposition (iCVD) technique. iCVD is an all-dry, solvent-free vapor deposition strategy for coating of polymers from their corresponding monomers. Use of initiator di-tert butyl peroxide (TBPO) together with the monomer precursors improved the deposition rates substantially while decreasing the activation energy required to carry out the surface polymerization reactions. Acrylic acid was chosen for its suitable polarity and ethyl acrylate was chosen for its hydrophobicity and fluidity. Furthermore, because of its low glass-transition temperature (Tg) EHA acts as a plasticizer in the copolymer film. P(EHA-co-AA) copolymers were synthesized using different ratio of EHA to AA flowrates in the reactor feed to achieve tunable adhesive properties. The adhesive properties of the obtained films were characterized by mechanical tensile tester. XPS, FTIR and SEM analyzes were performed to characterize the chemical structure, composition and surface morphology of the as-deposited nano-adhesive copolymer films. DSC analysis was used to reveal the Tg of the as-deposited copolymer films. It was found that highly homogeneous thin films of P(EHA-co-AA) can be obtained by iCVD with high retention of EHA and AA monomer functionalities. Very high deposition rates up to 400 nm/min were observed. As-deposited materials possessed high adhesive properties, which could be tuned by changing the comonomer ratios in the reactor feed.

Nanocomposite materials exhibit properties unreachable by conventional materials, as the nanoscale of the fillers allows for size-dependent effects including enhance mechanical and transport properties. However, these exact scales also complicate fabrication making traditional fabrication methods incompatible with ultra-high surface area and the van der Waals forces it creates. By combining chemical vapor deposition (CVD) with atomic layer deposition (ALD), bottom-up approaches, one can circumvent the difficulties plaguing the field of nanocomposites.
Alumina (Al₂O₃) exhibits excellent chemical and thermal stability and very good resistance to wear. However, low thermal and electrical conductance could be both an asset and a liability, sometimes in the same application. These properties can be tailored by incorporating highly thermally and electrically conductive carbon nanotubes (CNT). With that said, in order to conserve the anisotropic properties of CNT in the final form of the material, they must be aligned.
In this work, a vertically aligned carbon nanotubes (VACNT)/alumina nanocomposite material was fabricated. VACNT arrays were grown via water-assisted CVD and then coated with alumina via ALD. Before the ALD step, functionalization in the form of controlled oxidation was done in order to enhance the adhesion between the phases without destroying the anisotropicity. The synergistic effect of alumina and carbon nanotubes allows for improved thermal stability of the nanocomposite. Properties in the axial and transverse directions of the nanocomposite will be explained.

As strongly correlated oxides continue to be studied for next generation electronic devices, van der Waals epitaxy is gaining popularity as an effective technique to grow such oxides while ensuring a low-density of misfit and threading dislocations as well as minimum interdiffusion at the interface. In this work, we report an exception to this trend. While attempting to grow VO₂ via chemical vapor deposition on CsBiNb₂O₇, a layered Dion-Jacobson perovskite, we observe the epitaxial growth of CsVO₃ instead. This points to an aspect of van der Waals epitaxy not encountered so far - the possibility of an interaction between surface ions of van der Waals epitaxy substrates and the growing material. We confirm the growth of CsVO₃ by Raman mapping and reveal the epitaxial relation between CsVO₃ and CsBiNb₂O₇ via electron microscopy diffraction. Since CsVO₃ is an excellent broad-band emitting phosphor currently being explored for use in rare-earth free white light-emitting diodes, we also probe photoluminescence properties of the epilayer and model the probable electronic structure. Based on our study, we suggest that the substrate – oxide chemistry may need to be considerably taken into account while designing and implementing van der Waals epitaxy in correlated oxide systems.

With the purpose of increasing the efficiency and control of light emission from silicon quantum dots (SiQDs), several research groups have used different configurations of nanoparticles and nanostructures made by using noble metals (Au, Ag) near the silicon quantum dots ^[1,2]. A shared feature in the coupled structures reported in those previous works is that all of them have a spacer between the metal and the SiQDs. In turn, the spacer must be a dielectric and must have a well-defined thickness ranging from 10 to 20 nm.

In this work, we use nanolayered structures like those mentioned previously, to manufacture a light emitting device (LED) with a metal-insulator-semiconductor (MIS) configuration. SiQDs embedded in silicon nitride acted as luminescent films and pure silicon nitride serve as a dielectric spacer. Both films were deposited by Remote Plasma Enhanced Chemical Vapor Deposition (RPECVD). Meanwhile, gold nanoparticles (AuNPs) were deposited by sputtering on a p-type silicon substrate.

We evaluate the electroluminescent (EL) spectral enhancement factor which is defined as the ratio of EL intensities of samples with AuNPs and their references without AuNPs. In one of our samples, this ratio at 27.7 mA shows a maximum EL enhancement factor of 7 at about 510 nm and 5.4 at about the maximum EL intensity peak (~ 600 nm). It is worth to notice that the higher EL enhancement factor in these structures is close to the absorption peak of gold nanoparticles at 538 nm. Therefore, our devices allowed us to prove that there is indeed an electroluminescent enhancement when the appropriate AuNPs-spacer-SiQDs configuration is used.

We have found that EL enhancement may indeed be due to a plasmonic coupling. Nevertheless, we also identify that the presence of gold nanoparticles in the EL device allow a more efficient distribution of charge carriers towards the luminescent centers (SiQDs). Consequently, we conclude that more than one mechanism should be involved in the optimized electroluminescence.

We report on the influence of various nitrogen plasma species on the growth and structural properties of indium nitride in migration-enhanced remote-plasma metalorganic chemical vapor deposition. The ion energy and flux have been determined via the atomic emission spectroscopy by the analysis of emission spectra recorded at the growth surface. The atomic nitrogen ions’ flux has been found to have a significant effect on the growth rate as well as the crystalline quality of indium nitride films. Results of Raman spectroscopy, infrared reflectometry, atomic force microscopy, and charge transport Hall measurement are discussed.

Vanadium dioxide (VO₂) exhibits a unique low-temperature (~70 °C) phase-transition behavior due to a structural change from monoclinic phase (low temperature) to a tetragonal rutile phase structure (high-temperature). Moreover, this phase-transition is ultrafast (< ps) and reversible, which makes VO₂ attractive for a wide variety of applications. Also known as metal-insulator transition (MIT), this property can be effectively used in optical and electrical switches, RF-microwave switches, tuneable plasmonic and metamaterial systems, and smart windows. Among the synthesis methods used to deposit VO₂ films, chemical vapor deposition, reactive electron-beam evaporation, reactive magnetron sputtering, pulsed-laser deposition, sol-gel methods, and hydrothermal processes have been reported. Atomic layer deposition (ALD) of VO₂ films have relatively recently been of interest, mainly due to it’s precision thickness control, large-area uniformity, and low-temperature compatibility. Majority of VO₂ ALD research focused on thermal-ALD based growth recipes with either ozone or water vapor are used as the oxygen co-reactant and reported film properties as a function of post-deposition annealing.

In this work, we report on the low-temperature self-limiting growth of VO₂ thin films using tetrakis(ethylmethylamino)vanadium(IV) (TEMAV) as metal precursor and oxygen plasma as co-reactant in a plasma enhanced atomic layer deposition system. To the best of our knowledge, this is the first study on the plasma-ALD of VO₂ via oxygen plasma. Our main motivation behind using O₂ plasma is to reduce the necessary substrate temperature via energetic oxygen radicals and gaining additional degree of freedom in adjusting the film growth conditions through plasma variables such as plasma power, exposure time, O2 partial pressure and plasma gas composition.

VO2 films are deposited on Si(100) samples, which went through a conventional solvent cleaning process just before loading into the reactor chamber for the growth process. Ultratech Fiji G2.0 Plasma-ALD system has been used to carry out the deposition experiments. The substrate temperature has been set to a fixed temperature of 150 °C, mainly due to the narrow temperature window where TEMAV precursor has reasonable vapor pressure and is below the thermal decomposition temperature. The unit ALD cycle used for the VO₂ deposition consists of 30-100 ms TEMAV dose which is carried into the reactor via 20 sccm Ar-carrier flow, Ar-purge for 10 s, O₂ plasma (50 sccm) at 300W for 10 s, and finally another 10 s of Ar purge. In addition, the TMAV precursor is heated at 115 °C in order to provide sufficient amount of precursor vapor into reactor chamber. The total number of ALD cycles used for film deposition is 300. VO₂ film thickness is checked by ex-situ ellipsometry after deposition. The film thicknesses measured for TEMAV pulsing times of 30 and 100 ms are 15.5 and 20.5 nm, respectively. The extracted refractive index ranges within 2.25 - 2.35, while the growth per cycle (GPC) values varied from ~0.5 Å to 0.7 Å. The obtained results agree fairly well with previously published reports in the literature. X-ray diffraction measurements as a function of annealing temperature and ambient will be presented along with the detailed temperature-dependent electrical characteristics of the synthesized VO₂ films.

We report a comparative study investigating different atomic layer doping (ALDp) strategies to achieve effectively doped ZnO layers. The main motivation of our study is to gain insight into how the surface reactions are affected during the relatively conventional super-cycle approach and less-used co-dosing method. To achieve real-time information, we have carried out in-situ ellipsometric measurements which reveal sub-angstrom film thickness changes as a result of individual chemisorption, ligand exchange, and incorporation reactions. Aluminum-doped ZnO (AZO) has been studied as the material of interest, which finds a wide range of use in solar cells and transparent/flexible electronics as transparent conducting oxide layers.

Thermal atomic layer deposition (ALD) with diethylzinc (DEZ) and trimethylaluminum (TMA) were used as the metal precursors for ZnO film growth and Al-doping, respectively, while H2O was used as the common co-reactant. We experimented conventional sequential or super-cycle approach and simultaneous or co-dosing methods to form AZO layers with the target of reaching the highest possible conductivity. To the best of our knowledge, the co-dosing approach has not been studied or reported for AZO layers which motivated us for this systematic study. The deposition recipe used for AZO film growth is as the following: DEZ pulse or TMA pulsing for 15 ms; nitrogen purging at a flow of 20 sccm for 10 s; de-ionized water pulse of 15 ms as a co-reactant; nitrogen purging at a flow of 20 sccm for 10 s. The substrate temperature was varied as 100, 150, and 200 °C to understand its impact on doping efficiency. Total number of ALD cycles used is 300 cycles for each film, where the super-cycle ratios were tuned from 1:10 to 1:50 and the priority pulsing of DEZ and TMA is changed for co-dosing experiments.

We have observed that the refractive index of the AZO films changed between 1.95 - 1.87, depending on the Al-doping ratio for super-cycle doped AZO films. However, AZO films doped with the co-dosing approach resulted in films with a refractive index of around 1.6, which very much resembles Al₂O₃ films. To double-check this result, DEZ is first pulsed shortly before the TMA pulse, however the result did not change, resulting in an Al₂O₃-like refractive index. This observation indicates that chemisorbed DEZ groups are being replaced by incoming TMA molecules and as a result, ZnO is converted into Al₂O₃. We have carried out detailed in-situ analysis of the surface reactions for both doping approaches and particularly the co-dosing method where ZnO-to-Al2O3 conversion is observed. The ellipsometric data is supported by x-ray photoelectron spectroscopy (XPS) which reveals the chemical composition and level of atomic doping concentrations. The co-dosing approach is further tested for other group-III element precursors such as triethylgallium (TEG), trimethylindium (TMIn), and triethylboron (TEB), and these experimental results will be presented as well.

Nanoporous polymers are emerged as important elements due to their potential applications utilizing controllable optical, mechanical, and microfluidic properties, such as biosensor, which bulk polymer cannot offer. Parylene-C has been used for various micro-electromechanical systems (MEMS) and optoelectrical devices due to its electrical insulation and optical transparency. Moreover, it is commonly used as an encapsulation layer in biomedical devices to take advantages of biocompatibility, low water absorption rate, and pinhole free conformal coating.
Several methods were suggested to grow nanostructured parylene-C but still limited to a small area or difficult to control the porosity [1-2]. In this work, to overcome current limitations, we report the novel combinatorial approach of nanoporous parylene growth method by controlling process pressure and vapor flux. Commercial PDS2010 (Specialty Coating System) was modified with custom-designed 3D printed nozzles to provide the directional and localized vapor flux to the substrate, while other parameters are the same. With the elongated tip of nozzle geometry (rectangular) and in increased process pressure, the porous layers were formed due to the enhanced deposition rate. The uniform layers over the 4-inch diameter were successfully grown in the optimized nozzle design. Together with scanning electron microscopy observation, we also characterize the optical properties of the films with the spectroscopic ellipsometry tool for precise porosity measurement. Along with pressure- and flux controlled nanoporous polymer growth, potential applications and an outlook to the future will be discussed.

[1] Pursel, S., Horn, M. W., Demirel, M. C. & Lakhtakia, A. Growth of sculptured polymer submicronwire assemblies by vapor deposition. Polymer. 46, 9544–9548 (2005).
[2] Binh-Khiem, N., Matsumoto, K. & Shimoyama, I. Porous Parylene and effects of liquid on Parylene films deposited on liquid. IEEE 24th International Conference on Micro Electro Mechanical Systems 111–114 (2011).

The mist vapor deposition is one of the film formation technique of organic semiconductor materials that enables printed electronics. The liquid is ultrasonicated into particles of about 20 m in diameter, and sprayed onto the heated substrate. The evaporation rate of the solvent is controlled by the substrate temperature. At this time, innumerable micro droplet traces, that is, coffee rings are formed. In the solution process, this undesired structure are affecting the physical and electric properties of formed films. Similarly, in the spray method and the ink jet method, the formation of this coffee ring is one of the causes of deterioration in film quality as compared with vacuum deposition. We examined controlling the generation of micro-coffee ring in the mist deposition method by changing the evaporation rate of the solvent.
A bare fullerene was used as the organic semiconductor material. Two solvents having different boiling points were mixed and used as a solvent. A solvent with a high boiling point is A, a solvent with a low boiling point is B, and the mixing ratio is 20%, 25%, 30%. The difference between the boiling points of these solvents was about 80 degree. The substrate temperature during film formation was changed in the range of 120 to 300 degree. The surface morphology was observed by AFM, and ultraviolet spectroscopy (UPS) and XRD measurements were performed.
When the substrate temperature was high a uniform coffee ring with a diameter of about 15-20 microns was formed at 300 degree regardless of the mixing ratio of the solvents. The surface roughness Rq was about 35 nm to 45 nm. On the other hand, when the film formation temperature was changed in the range of 180 to 210 degree, Rq was 7nm to 20 nm. Significant improvement of surface roughness, and no formation of coffee ring was observed in the solvent ratio of 30%.
A variation was found in the HOMO level determined by the UPS measurement results due to the difference between the solvent mixing ratio and the film formation temperature. The HOMO level changed in the range of 5.7 eV to 6.26eV. The value of vacuum evaporated fullerene thin film is known to be 6.25 eV. When the mixing ratio of the high boiling point solvent was 25% and 30%, it showed a lower value than the HOM level of vacuum evaporation. On the other hand, when the mixture ratio of the solvent was 20%, the HOMO level equivalent to vacuum evaporation was shown. In particular, the surface roughness was low and the value was at the same level.
We controlled the generation of coffee ring and realized a thin film with the same roughness as vacuum deposition by changing the mixing ratio using high boiling point solvent and low boiling point solvent. Furthermore, by controlling the surface roughness, a thin film with a HOMO level equivalent to that of vacuum deposition was achieved.

We describe the atomic layer deposition (ALD) of titanium oxysulfide films. A new ALD chemistry of tetrakis (dimethylamido) titanium (IV) and hydrogen sulfide (H₂S) is suggested for the synthesis of amorphous titanium sulfide layers. We found that the resulting films subsequently underwent oxidation upon reactions under ambient condition, resulting in titanium oxysulfide (TiO_2-xS_x). The resultant structures were analyzed by using X-ray diffraction, transmission electron microscopy, and X-ray photoelectron spectroscopy, indicative of the formation of TiO_2-xS_x. A combined study of Hall-effect measurements and Mott-Schottky analysis showed n-type semiconductor behaviors possessing a good conductivity. Optical properties testify that the present system has a moderate band gap in between the related binary end compounds such as TiS₂ and TiO₂

Three dimensional (3D) encapsulation of free-standing particles and powders by atomic layer deposition (ALD) remains a technical challenge. Current solutions, such as fluidized bed reactors, rotary tumblers etc, employ physical agitation to vary the orientation of the particle during the deposition process, thereby exposing all surfaces to the flowing precursors. As a result, these methods require large amounts of the material to be coated, with minimum working quantities typically on the order of 10s to 100s of mg. A technique for encapsulating static particles on substrates would enable the processing of small volumes of material (c.f. bulk powders), down to individual micro- and nano-particles.

Here, we demonstrate the coating on all sides of individual particles resting on an inert polystyrene film. Vapor-phase infiltration [1] is performed on a commercial ALD tool. During this process, organometallic precursors diffuse through the inert polymer and lead to the growth of a target inorganic film on the particle’s surface, even on the bottom side that is in contact with the polystyrene. We examine the uniformity of the coating by cross-sectional transmission electron microscopy and measure its thickness for different numbers of infiltration cycles.

To demonstrate the usefulness of our technique, we coat III-V semiconductor microdisk particles with a thin layer of aluminum oxide. These microdisks support whispering gallery mode resonances and under optical pumping exhibit single-mode lasing. As free-standing particles, microdisk lasers have applications in biomedical imaging as intra-cellular probes [2]. We show that a thin (sub-10-nm) coating of aluminum oxide applied by our technique can improve the stability of lasing emission under continuous illumination in an aqueous environment compared with uncoated semiconductor disks.

Our technique is applicable to a range of particles, and generalizable to the growth of different inorganic films. Moreover, sequential vapor infiltration can be performed on a regular ALD tool without needing a special chamber or cleaning procedure.

[1] A. Subramanian, A., Tiwale, N., & Nam, C. Y. (2019). Review of Recent Advances in Applications of Vapor-Phase Material Infiltration Based on Atomic Layer Deposition. JOM, 71(1), 185-196.
[2] Martino, N., Kwok, S. J., Liapis, A. C., Forward, S., Jang, H., Kim, H. M., ... & Lee, Y. H. (2018). Wavelength-encoded laser particles for massively-multiplexed cell tagging. Nature Photonics [in press], bioRxiv, 465104.

In this work, we have investigated the influence of critical growth parameters on the electrical and optical properties of indium nitride (InN) thin films grown at relatively low-temperatures via hollow-cathode plasma-assisted atomic layer deposition (HCPA-ALD). InN films were deposited on Si(100) substrates using trimethyl-indium (TMI) and variants of Ar/N₂/H₂plasma (N₂-only, Ar/N₂, and Ar/N₂/H₂) as the metal precursor and nitrogen co-reactant, respectively. ALD growth experiments have been performed within 50 - 150 W plasma power range and 160 - 240 °C substrate temperature. Dynamic in-situ ellipsometry was employed to observe individual ligand exchange events in real-time during the growth process. Additionally, ex-situ characterizations were done to identify the optical, structural, and chemical properties of the grown InN films. X-ray diffraction (XRD) results showed that only the samples grown at 200 and 240 °C with 100 W rf-power displayed single-phase hexagonal crystalline structure with peak intensity values increasing as a function of substrate temperature. Moreover, varying the plasma chemistry such as addition of H₂ to Ar/N₂ led to significant microstructural changes resulting in crystal quality degradation and porous films with high carbon content. As such, x-ray photoelectron spectroscopy (XPS) measurements are carried out to further understand the possible reactions taking place in the presence of H₂ plasma. The electrical properties (resistivity, carrier density, and Hall mobility) of the polycrystalline InN films are analyzed using I-V and Hall measurements and will be presented in detail. Spectroscopic ellipsometer is utilized to study the spectral absorption characteristics of the optimal InN films to extract the corresponding optical band edge. Finally, with the highest quality InN films, a proof-of-concept visible/near-IR photodetector will be fabricated and tested for performance.

Germanium (Ge) is an impending p-channel substrate to overcome the drain-current saturation problem of MOSFET due to its higher carrier mobility. However, the most critical challenge for adopting Ge as the substrate for the next generation p-MOSFET is the instability at high-k/Ge interfacial region. Strontium titanate, SrTiO₃(STO) film was selected as the dielectric because its bulk dielectric constant is as high as 300. In this study, therefore, the growth behaviors and electrical properties of the atomic layer deposited (ALD) S_rTi0₃ were studied with the various deposition temperature and annealing temperature. SrTiO₃ films were deposited by using Sr(iPr₃Cp)₂ and Ti(CpMe₅)(OMe)₃ (Pr, Cp, and Me are propyl, cyclopentadienyl, and methyl groups, respectively) precursor on Ge substrates at 230°C and 370°C . 370°C deposition showed more initial excess deposition of Sr element than 230°C deposition. Also, a higher growth rate was confirmed at the higher deposition temperature. For ex-situ crystallization, the STO films were annealed at the various annealing temperature ranging from 550-650°C for 5min by rapid thermal annealing. For the electrical analysis, Metal-Insulator-Semiconductor (MIS) capacitors were prepared by depositing TiN/Pt as the metal gate. An extremely high dielectric constant (~285) was confirmed from the slope of the equivalent oxide thickness (EOT) vs. physical oxide thickness (POT) plots. However, the quite high interfacial oxide thickness increased the total EOT values, requiring further interfacial engineering.

Recently, there were many researches on two-dimensional (2D) materials having semiconducting characteristics like transition metal dichalcogenides (TMDCs). These researches have mainly focused on the realization of original 2D materials characteristics through deposition process. However, representative TMDCs such as MoS₂ and WS₂ have tried many different disposition methods to overcome the limitations of their high process temperature (≥750°C). Due to their high process temperature, these materials can not be directly applied onto flexible polymer substrates. On the contrary, Tin disulfide (SnS₂) is post transition metal sulfide and has low process temperature (≤350°C), which is low enough to directly deposit on polymer substrates, and exhibits comparable characteristics with MoS₂ and WS₂ because of similar hexagonal layered structure. Therefore, we have developed low temperature process of few layer SnS₂ thin films with atomic layer deposition method for obtaining 2D material characteristics and investigated the electrical properties by fabricating thin film transistors (TFTs).
In this study, SnS₂ thin films were grown by atomic layer deposition with tetrakis(dimethylamino)tin (Sn[N(CH₃)₂]₄, TDMASn) and hydrogen sulfide (H₂S, 99.9%) at 150 °C. Subsequently, the post deposition annealing process up to 350°C was performed in 4% H₂S gas ambient. We then identified the crystal structure and chemical bonding using X-ray diffractometer (XRD) and X-ray photoelectron spectroscopy (XPS), and the number of SnS₂ layers was examined by atomic force microscopy(AFM), Raman analysis and high resolution-transmission electron microscopy (HRTEM). To investigate the switching characteristics of few layer SnS₂ thin films, we fabricated bottom gate TFTs using few layer SnS₂ thin films as active channel layer. As a result, few layer SnS₂ TFTs showed n-type characteristics with on/off current ratio 8.32 x 10⁶, mobility 0.08 cm²/Vs and 830 mV/decade of subthreshold swing.

The appeal of microwave-assisted atmospheric plasma for surface modification and embedding of chemistry is thwarted by the lack of predictive link between the process parameters, plasma chemistry, surface plasma reactions, and growth of functional layers with control over morphology and phase. The atmospheric plasma deposition of zirconia on a natively grown aluminum oxide layer offers an advanced, robust and cost-effective processing without a vacuum chamber in ambient air. Here, we present a comprehensive multiphysics modeling of the plasma deposition to study the physics of the processing and the metal oxide growth on the substrate. An organometallic evaporation model is adapted for predicting evaporation rate of precursor. This model calculates the growth rates and film thickness of zirconia and demonstrates strong dependency among growth rates and processing parameters including flow rates at different inlets and torch-substrate distance. This optimization study can help to enhance the performance of the experiments and achieve higher growth rate of metal oxide on substrate.

We report the nature of reverse leakage current characteristics in InGaN/GaN light emitting diodes (LEDs) on freestanding GaN (FS-GaN) crystals detached from a Si substrate for the first time, using temperature-dependent current-voltage (T-I-V) measurement
Based on T-I-V measurement results, carrier transport mechanism of the reverse leakage current is analyzed in the InGaN/GaN LEDs. Their conduction mechanism can be divided into variable-range hopping and nearest neighbor hopping (NNH) around 360 K, which is enhanced by Poole-Frenkel emission. The analysis of T-I-V curves of the homoepitaxial LEDs yields an activation energy of carriers of 35 meV at -10 V, about 50% higher than that of the conventional ones (Ea = 21 meV at -10 V). This suggests that our homoepitaxial InGaN/GaN LEDs bears the high activation energy as well as low threading dislocation density (about 1 × 106/cm2), effectively suppressing the reverse leakage current. We expect that this study will shed a light on the high reliability and carrier tunneling characteristics of the homoepitaxial InGaN/GaN LEDs produced from a Si substrate and also envision a promising future for their successful adoption by MEMS community.

Significant development of high luminescence efficiency in InGaN/GaN LEDs has offered new futuristic applications such as automotive headlamps, traffic signals, displays, and general lighting. Despite their rapid advances and successful applications in lighting sources, improved efficiency, functionality, and flexibility in LEDs must still be obtained to use LEDs in emerging industries such as the internet of things (IoT), bio-applications, and wearable devices. Furthermore, integration of GaN and Si process enables tremendous new possibilities to Si-based MEMS and system designers.

In this work, we demonstrated homoepitaxial InGaN/GaN LEDs on FS-GaN substrate, which was grown by metal-organic chemical vapor deposition and detached from a Si substrate. To compare the reverse leakage current characteristics in InGaN/GaN LEDs fabricated using FS-GaN grown from a Si substrate, conventional InGaN/GaN LEDs with corresponding structures and the peak emission wavelength were fabricated on sapphire substrate. All the LED structures were fabricated with conventional lateral chip process. We performed temperature-dependent reverse leakage current measurement of InGaN/GaN LEDs. Based on T-I-V measurement results, we will discuss carrier tunneling characteristics of the homoepitaxial InGaN/GaN blue LEDs produced from a Si substrate.

This talk will discuss the mechanism, kinetics, and applications of initiated chemical vapor deposition onto liquid substrates such as silicone oils and ionic liquids. We will discuss how the surface tension interaction between the liquid and polymer governs the dynamics at the vapor-liquid interface leading the formation of either polymer particles or polymer thin films. We will demonstrate that the viscosity of the liquid also affects the dynamics at the vapor-liquid interface by affecting the diffusion of the polymer chains. For some systems, the monomer can absorb into the liquid substrate. In these cases, polymerization can occur within the bulk liquid leading to the formation of gels. We will also show that we can combine initiated chemical vapor deposition with sputtering onto liquid substrates to form hybrid inorganic-organic materials.

Since its discovery, Poly(3,4-ethylenedioxythiophene) (PEDOT) has emerged as one of the most popular intrinsically conductive polymers. Oxidized PEDOT’s transparency, electrical conductivity, thermal and electrochemical stability, biocompatibility, and thermoelectric properties make it suitable for a wide range of applications which include smart textiles, photovoltaic devices, prosthetics, and capacitors.
PEDOT’s rigid conjugated backbone allows for the formation of crystalline domains with short inter-chain distances and the formation of bipolaron and polaron networks, corresponding to semi-metallic and metallic character, respectively. Although beneficial to performance, the rigid conjugated bond structure of oxidized PEDOT decreases its processability due to increased thermal stability and insolubility. In order to overcome these issues, PEDOT formed with the hydrophilic stabilizing anion Poly(styrenesulfonic acid) (PSS) has been widely used as a solution processable form of PEDOT. However, PSS- stabilizing anions have recently been correlated to anomalous currents in electrolytic capacitors, anisotropic conductivity, and decreased crystallinity. In addition, slurries of pre-polymerized PEDOT limit its infiltration and growth in high aspect ratio structures. Vapor-phase deposition processes are therefore promising methods to increase PEDOT film performance and applications.
Until recently, vapor phase synthesis of Poly(3,4-ethylenedioxythiophene) (PEDOT) has relied on solid oxidants which require complex heating and dosing schemes for sequential layer-by-layer processes. Until now, vapor-phase PEDOT deposition has therefore been limited to to vapor phase polymerization (VPP) and oxidative chemical vapor deposition (oCVD) processes.
This work introduces a novel oxidative molecular layer deposition (oMLD) process using the volatile liquid oxidant, antimony(V) chloride (SbCl₅), to deposit high performance PEDOT thin films with nanometer precision. Using a home built, hot wall, viscous flow reactor, semi-metallic PEDOT thin films have been deposited with record breaking thin film conductivity of 6700 S cm^-1 at moderate deposition temperatures (150 ^oC). In addition to producing high performance PEDOT thin films, the protocol developed in this work has been used to explore the effects of reactor parameters on PEDOT nucleation and growth as well as film characteristics, moving towards greater control over PEDOT performance. Furthermore, this facile oMLD process has enabled the study of substrate dependent nucleation and growth, thereby moving towards an enhanced understanding of PEDOT selective deposition.

Impressive values and systematic tunability of the optoelectronic properties of poly(3,4-ethylenedioxythiophene) (PEDOT) are demonstrated using single-step all-dry process. With the volatile oxidant, vanadium oxytrichloride (VOCl₃), pure face-on orientation results from oxidative Chemical Vapor Deposition (oCVD) without any post-deposition rinsing. In contrast, predominately edge-on orientation previously resulted with the solid oxidant FeCl₃ at the similar oCVD process conditions and a post-deposition rinsing step is required. At growth temperatures compatible with direct deposition on plastic substrates (140 °C), using the single-step oCVD process with the volatile VOCl₃ oxidant, we obtained ~four-fold improvement of direct current conductivity to optical conductivity ratio ) and ten-fold improvement of electrical conductivity as compared to oCVD PEDOT produced using FeCl₃. Indeed, the maximum = 50 for this mechanically flexible polymer exceeds the benchmark for commercial transparent conductors. Varying only the partial pressure of the VOCl₃ oxidant induces systematic variation of the b-axis lattice parameter in the oCVD PEDOT crystallites from 6.98 to 7.02 Å and 6.97 to 7.01 Å for films grown at the deposition temperature of 110 °C and 140 °C, respectively. The lowest value correlates with the highest electrical conductivity, largest optical band gap (2.9 eV), and lowest degree of disorder as characterized by the Urbach band edge energy. The utility of the optimized oCVD PEDOT is demonstrated as a hole transport layer (HTL) for the fabrication of an inverted perovskite solar cell (PSC). This device achieves a power conversion efficiency (PCE) of 18.04%, higher than the 16.20% PCE for the control structure having a spun-cast PEDOT:PSS HTL. More significantly, incorporating oCVD PEDOT, rather than PEDOT:PSS, increased device stability approximated two-fold over a 42-day evaluation period.

Stimuli-responsive materials are characterized by dynamic switching of their properties depending on external stimuli (e.g. light, pH, temperature, humidity). In particular, hydrogels change their size and shape when exposed to aqueous environments. Functional and responsive surfaces have been successfully deposited by initiated Chemical Vapor Deposition (iCVD) on a variety of substrates. 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 setup was designed without electric components in the vicinity of the active sensor layer and is therefore applicable in harsh environments such as explosive or corrosive ones. 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.

Another case of study will be presented in the field of multi-stimuli-responsive materials. A chemical functionalization of the hydrogel surface was performed to add multiple stimuli-responsive functionalities and obtain a smart material that responds to two stimuli at the same time. Modifying the hydrogel surface with solution-based methods is often problematic because of the damages caused by the permeation of solvents in the hydrogel. This issue is completely bypassed by the use of solvent-free techniques, like iCVD. Such polymers were used as drug encapsulants to achieve controlled drug release upon stimuli, with possible application as wound dressings.

Initiated chemical vapor deposition (iCVD) is a solvent-free, cost efficient synthesis technique to deposit conformal polymeric thin films. The properties of these organic films can be precisely modified by the deposition parameters. The use of different comonomers facilitates copolymerization, which for instance, enables further control over the crosslinking degree and film functionality. This opens a wide field of possible applications, like surface functionalization, functional dielectrics and drug delivery systems. Here, we demonstrate the tailored synthesis of hydrogel-, organosilicon- as well as fluoropolymers and their respective applications for conformal coatings on porous 3D structures, superior functional electret layers, photoswitchable copolymer films, drug delivery systems with barrier layers and adhesion promotion by use of various comonomers within one iCVD process.

The success of graphene opened a door for a new class of chalcogenide materials with unique properties that can be applied in the semiconductor technology [1]. Monolayers of two-dimensional transition metal dichalcogenides (2D TMDCs) possess a direct band gap [2] that is crucial for optoelectronic applications. Additionally, the direct band gap can be easily tuned by either chemical composition or external stimuli. Next to the optoelectronic applications, where a monolayer planar structure is necessary to employ, a layer of standing flakes, which possesses a large surface area, can be used for hydrogen evolution [3] a photodegradation of organic dyes [4] or as electrodes in Li ion batteries [5].
In principle, TMDCs can be prepared by various top-down (e.g. exfoliation) and bottom-up techniques, such as chemical vapor deposition (CVD) and atomic layer deposition (ALD) growth techniques [1]. MoS₂, a typical representative of TMDCs, has been widely studied for many applications. Recently, the possibility to employ ALD as a technique to grow MoS₂ has been reported. In these works (CH₃)₂S₂ [6] or H₂S [7, 8] were used as the S precursor and Mo(CO)₆ [6], MoCl₅ [7] or Mo(thd)₃ [8] as the Mo precursors. From the practical point of view, MoSe₂ is even more interesting than MoS₂ since MoSe₂ possesses a higher electrical conductivity than MoS₂ [9, 10]. Recently, we have shown that ALD deposition of MoSe₂ [11] or Mo-O-Se [12] using ((CH3)3Si)2Se as the Se precursor and the MoCl₅ or Mo(CO)₆, respectively, as the Mo precursors is feasible.
The presentation will focus on the synthesis of MoS₂ and MoSe₂ by ALD, their characterization and applications in various fields. Experimental details and some recent photocatalytic and hydrogen evolution results will be presented and discussed.

References:
[1] A. V. Kolobov, J. Tominaga, Two-Dimensional Transition-Metal, Dichalcogenides. Springer Series in Materials Science, Springer International Publishing AG, Switzerland 2016
[2] B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti, A. Kis, Nat. Nanotechnol. 2011, 6, 147.
[3] L. Wang, Z. Sofer, J. Luxa, M. Pumera, Adv. Mater. Interfaces 2015, 2, 1500041
[4] Y. Wu, M. Xu, X. Chen, S. Yang, H. Wu, J. Pan, X. Xiong, Nanoscale 2016, 8, 440
[5] D. Ilic, K. Wiesener, W. Schneider, H. Oppermann, G. Krabbes, J.Power Sources 1985, 14, 223
[6] Z. Jin, S.Shin,D.H.Kwon, S. J.Han,Y. S.Min,Nanoscale 2014, 6, 14453.
[7] L. K. Tan, B. Liu, J. H. Teng, S. Guo, H. Y. Low, K. P. Loh, Nanoscale 2014, 6, 10584
[8] M. Mattinen et al., Adv. Mater. Interfaces 2017, 4, 1700123.
[9] D. Kong, H. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao, Y. Cui, Nano Lett. 2013, 13, 1341.
[10] A. Eftekhari, Appl. Mater. Today 2017, 8.
[11] M. Krbal et al., Phys. Stat. Sol. RRL, 2018, 12, 1800023
[12] S. Ng et al., Adv. Mater. Interfaces 2017, 1701146.

Two dimensional (2D) materials have reshaped the research interest in the field of electronics and aspired the researchers to develop ultrathin optoelectronic devices. However, in order to steer the progress on a fast track, the synthesis of high-quality 2D materials with desired architect is imperative. In this work we have demonstrated a precursor “gettering” approach for chemical vapor deposition (CVD) to directly grow patterned graphene as well as hexagonal boron nitride (hBN) on Cu substrate using a predesigned mask. Our proposed method is advantageous to the previous reports, which generally involved several post-growth complicated lithographic steps to obtain the desired architect of graphene films. Moreover, the quality of films grown via our method comprises of high quality in terms of layer homogeneity. Interestingly, taking advantage of the same growth substrate (Cu), we have constructed lateral heterostructures of graphene and h-BN through sequential feeding of respective precursors, in one-step CVD process. However, lateral heterostructures growth have been demonstrated previously but those are rather randomly distributed or others require multiple growth steps and complicated lithographic route. Therefore, our method is providing a unique strategy of one-step facile CVD process to build such patterned growth of graphene arrays and their heterostructures that can be integrated with crafting the nanoelectronic circuitry.

Due to the amazing and diverse properties of two dimensional (2D) materials, including semiconducting, metallic, insulating, and magnetic, there has been a great interest for more than a decade. A commonality among this class of layered materials is their ability to be thinned and stabilized down to one layer thick. The ability to achieve this thickness with minimal property degradation or even enhancement has generated continued interest for numerous applications. In these atomically thin materials, interactions with the environment, including the substrate and other 3D device elements, greatly degrade properties. Scattering from charged impurities, roughness, and phonons will reduce transport properties affecting device performance. The best solution has been the use of a thin van der Walls (vdW) buffer layers, like hBN. Of the available vdW materials, hBN is one of the only insulators, and due to is dielectric properties, high surface optical phonon energy, and atomically smooth surface, it has been the most successful and widely used material for preserving and protecting the intrinsic properties of 2D materials (graphene, phosphorene, MoS2, …). Here, we investigate the application of few-layer sp²-BN as a weakly interacting substrate, passivation layer and dielectric for 2D devices. By metal organic chemical vapor deposition (MOCVD), we grow few layer wafer scale (2”) sp²-BN on sapphire with excellent uniformity. To study this materials application as a weakly interacting encapsulation layer, we use transferred graphene to fabricate arrays of field effect transistors (FET) and investigate impacts on material and device performance compared to conventional dielectric substrates. As a substrate for graphene, MOCVD BN on sapphire results in graphene with a more that 2x increase in mobility and 10x reduction in sheet carrier density compared to on sapphire due to the lower interface doping and SO phonon energy. Through control of MOCVD process conditions, including temperature, pressure and V/III ratio, we can vary BN thickness and surface morphology, which we have used to investigate the role of BN morphology on transport properties. We also explore BN as a passivation layer for protection of conducting 2D layers from the ambient environment and oxidation. Finally, we discuss the use of BN in a 2 layer dielectric stack with atomic layer deposition Al₂O₃ allowing us to form high quality gates without the need for Al seeding.

The controllable, reproducible and scalable growth of graphene and related 2D materials remains the foremost challenge for both research and any technologies exploiting their unique properties. While chemical vapour deposition (CVD) has become the most widespread method for 2D material film growth, even basic process parameters remain not well understood due to the manifold, complex parameter space. Here we focus on widely used Cu as process catalyst and reveal the role and control of residual bulk oxygen, as well as showing new integrated process pathways, for achieving uniform Cu(111) epitaxial layers.
While oxidation is widely used to remove impurities in metal catalysts and to control the nucleation density of graphene[1], we show that minute concentrations of residual bulk oxygen can significantly deteriorate the quality of as-grown graphene highlighted by an increased Raman D/G ratio, increased propensity to post-growth etching and increased fraction of multi-layer graphene nucleation[2]. Starting from commercial Cu foils, we show that a simple hydrogen annealing step after the initial oxidation allows us to lower the residual oxygen level as measured by time-of-flight secondary ion mass spectrometry to produce graphene of significantly higher quality. This can be effectively combined with a short hydrocarbon exposure time of 10 min to achieve near full mono-layer graphene coverage, suitable for emerging industrial applications. We show that residual oxygen can have an equally significant impact on Fe catalysed h-BN CVD[2,3], and discuss the underlying mechanisms with parallels to well-known processes in metallurgy, catalysis and vacuum science. We further demonstrate the scalable deposition of ultra-low roughness single crystal Cu(111) and 2D material growth within a single synthesis step. The total synthesis from catalyst to 2D film takes less than 2 hours, during which there is no exposure to atmospheric conditions, circumventing the most common source of contaminants. A matching rapid and scalable method of transfer is devised and further improvements for device integration will be discussed[4,5].

References
[1] Braeuninger, P. et al. (2016). Chem. Mat. 28.24, pp. 8905–8915
[2] Burton, O.J. et al. (2019) J. Phys. Chem C. (Just Accepted)
[3]Caneva, S. et al. (2016) Nano Lett., 16 (2), 1250–1261.
[4] Wang, R. et al.(2019). ACS Nano. 13 (2), pp 2114–2126.
[5] D. De Fazio et al. (2019). arXiv:1904.01405

Since 2011, 2D semiconductor monolayers such as transition metal dichalcogenides (TMD including MoS2, WS2, MoSe2, WSe2, etc), have attracted tremendous attention owing to their extraordinary properties, setting the stage for new breakthroughs in fundamental nanoscience and applications, from electronics to biosensors. To date, various techniques have been explored to obtain TMD monolayers, and among them, chemical vapor deposition (CVD) synthesis using transition metal oxide and chalcogenide solid precursors is the most commonly adopted approach in many laboratories. In particular, the transition metal oxide precursor was placed at the center of a furnace while the chalcogenide precursor at the upstream of the furnace, with targeted substrate facing down and put above the crucible containing transition metal oxide precursor. Notably, the amount of precursor used in previous reports is normally surplus, which very often gives rise to chemical reactions between the precursors in each of their containers during synthesis, due to the diffusion of both precursors at the growth temperature. Thus, typically one growth chamber is dedicated to the growth of only one type of TMD to avoid cross-contamination (unless the chamber is used for hetero-TMD structures growth). In another words, the number of TMD that can be synthesized in a lab is limited by the number of CVD setup available in the lab.

We addressed this challenge by redesigning the CVD setup and pruning the amount of precursor used. We have been able to simultaneously synthesize different types of TMD on separate substrates using the same growth chamber. Through computational fluid dynamics modelling, we found that gas species trapped within those slanting substrates in our CVD setup are unable to escape from the trap, and thus avoiding the cross-contamination issue. The synthesized TMD films exhibit high-quality as confirmed by Raman, PL, XPS, AFM, and STEM analyses. Field-effect transistors fabricated on our TMD monolayers exhibit high electron mobility (64 cm²/V-s for MoS₂ and 21 cm²/V-s for WS₂ at room temperature) and large on-off current ratio (10⁷). Remarkably, our devised CVD technique has two advantages: (1) a CVD setup can be used to grow multiple TMD monolayers, which saves a lot of cost and space; (2) Multiple types of TMDs can be synthesized on separate substrates in one CVD cycle.

We also proposed a two-step CVD strategy to construct TMD-only synthetic electronics that can perform logic operations. The backbone of synthetic electronics relies on the possibility to chemically synthesize heterogeneous junctions between conductor and semiconductor, with perfect lattice matching. Interestingly, we found that our one-dimensional, solely TMD made, conductor-semiconductor junctions with large lattice mismatch exhibiting good electrical transport behaviors. We measured contact resistance as low as 500 Ωµm and a Schottky barrier height as small as 30 meV, both among the best values reported to date for contacts to 2D TMDs. Through a combination studies of STEM and second harmonic generation imaging, we confirmed that the semiconductor TMD monolayers nucleate from the vertexes of multi-layered TMD conductor and evolve into monolayer polycrystals. This is an unconventional non-edge epitaxy growth mechanism, and further studies are conducted to understand the interface physics of these lateral TMD junctions. In short, our CVD strategy enables the fabrication of ultrathin, flexible electronics, without the use of complex and costly device fabrication processes.

References:
[1] Leong et al., JACS 140, 12354–12358 (2018).
[2] Leong et al., Adv Funct Mater 27, 1605896 (2017).
[3] Leong et al., Nature Communications 10, 867 (2019).

**Appreciate if this can be indicated as an invited talk. Thank you.

Two dimensional materials namely graphene, MoS₂ and borophene went into focus of scientists all over the world due to their interesting properties (e.g. conductivity, flexibility) and possibility of large scale processability. However, the fabrication is bound by the surface chemistry and geometry of the used substrates, as well as the chemical interaction (adsorption and desorption) between substrate and precursor. In this regard, the formation of 2D materials on hierarchical superstructures is very difficult to achieve. Plasma enhanced CVD (PECVD); however, offers a way to overcome the chemical and geometrical restrictions, since the technique is not only able to activate the surface through heavy ion bombardment, but furthermore offers far-from-equilibrium phase formation at room temperature without substrate restrictions.
Here, we report the formation of three-dimensional, hollow h-BN superstructures, as obtained from a combined PE- and MOCVD deposition of B-N bond carrying single-source precursor molecules on ZnO host structures. Through the combination of plasma power, process temperature and time, we were able systematically study h-BN growth on these host materials to clearly demonstrate the suitability of the process for the formation of three-dimensional superstructures from two dimensional materials.

Metal nitrides films stand out as candidates for many strategic industrial applications as they exhibit superior functional properties such as mechanical, electrical and thermal properties. Complementary chemical vapor deposition techniques from High Temperature Chemical Vapor Deposition (HTCVD) to Thermal and Plasma Enhanced Atomic Layer Deposition (T-ALD and PEALD) have been investigated to fabricate metal nitrides thin films. In each case, one of the major challenges is the synthesis of high quality, pure (with no oxygen contamination) material. Among ALD developments, efforts have been focused on the exploration of different precursor molecules, chemical reactions as well as growth processes sequences and conditions.
In this presentation, the examples of aluminum nitride (AlN) and niobium nitride (NbN) deposition process development will be presented. Aluminum nitride (AlN) is a semiconductor material with a wide bandgap, excellent thermal conductivity, high electrical resistance and good behavior towards oxidization and abrasion. AlN films are attractive for many applications in energy, electronic or optoelectronic devices. For instance, thin films are used in piezoelectric transducers and microwave filters applications, in resonators, as passivating coatings, as AlN substrate in high power applications. NbN films are mainly used for single-photon detectors (SSPDs), superconductive radio frequency (SRF) cavities due to their superconducting properties.
We report on the optimizing routes and strategies to obtain the best compromised film properties on these two systems.

Increasing the energy that is coupled into a plasma by adding a magnetic field of 875G to it to make it ignite at the electron cyclotron resonance- (ECR) frequency (2.45 GHz) changes the gas-phase chemistry that occurs in standard plasma-assisted CVD processes. Films are formed as gas-phase reaction products at significantly higher ion energies (15–30 eV) and, at substrate temperatures that are appreciably lower than those that can be achieved without the electron-cyclotron augmentation. This phenomenon greatly affects the films’ properties.

In this study, the plasma ECR-MW-CVD technique is applied to make different types of hydrocarbon films, of variety of sp³, sp², and sp content, and densities (1.1 to 3.2 g/cm³), at substrate temperatures that range from 30 to 200^oC. The role that the composition of the gases (methane, hydrogen, and argon) in conjunction to the plasma power has on the films’ properties is discussed in the context of the enhanced gas-phase chemistry that occurs in the ECR-MW environment. The elemental composition of the films is determined from FTIR and Raman. Their surface morphology, in particular: their smoothness and texture, which are critical for the application that uses these materials as ablators for laser fusion experiments, are observed using SEM and AFM. Other films’ properties that are affected by the energetic species in the ECR-MW environment are determined via gravimetry, nanoindentation, and ellipsometry. The presentation shows the relationship that exists between the chemistry happening near the gas/substrate interface within the ECR-MW environment and the films’ properties.

This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0003856, the University of Rochester, and the New York State Energy Research and Development Authority.

The present work demonstrates a novel approach for the conformal deposition of ultra-thin conductive silver (Ag) films on complex substrates. Using an original multi-step plasma-enhanced approach, we demonstrate the deposition of conductive silver films with thickness down to 11 nm. Conductive ultra-thin Ag films are commonly deposited by physical vapour deposition techniques (metal evaporation or sputtering). However, these are line-in-sight methods that do not allow conformal deposition on substrate with complex morphology (e.g. trenches) [1]. Non line-in-sight methods such as the ones based on chemical vapour phase deposition (Chemical Vapour Deposition-CVD or Atomic Layer Deposition-ALD) usually produce non-electrically-conductive films for thickness below 20-50 nm due to island formation by Volmer-Weber growth mode for metal layers on oxide-based substrate [2]. Moreover, conductive ultra-thin silver films based on ALD (down to 22 nm) [3], require low-temperature deposition (below 120°C), with a very narrow process temperature window (about 10°C).

The deposition is based on Ag(fod)(PEt₃) as silver precursor, and H₂ plasma as reducing agent. An acceptable process temperature window (> 30°C), and a deposition temperature larger than 120°C for enhanced uniformity and deposition rate have been demonstrated. In these conditions, highly uniform deposition on bent and batched samples have been performed. 20 nm thick films exhibit a resistivity down to 40 μΩ.cm. The resistivity increases up to 1.4 mΩ.cm when reducing the thickness down to 11 nm. The critical thickness below 11 nm is thus very close to the state-of-the-art for sputter-deposited thin films [4], and well below the 22 nm obtained with ALD [3]. The obtained Ag thin film also demonstrates a high reflectance up to 94% and a low absorbance of 3% in the infrared region for a film thickness of 26 nm, showing the high quality of the films. This quality is confirmed by morphological analysis using scanning electron microscopy and atomic force microscopy, as well as by the structural (X-ray diffraction) and chemical (energy dispersive X-ray spectroscopy) properties. This new processing approach opens a very interesting path for the use of ultra-thin silver films for electronic and optoelectronic applications.

[1] Guske et al., Optical Society of America, 20, 21 (2012)
[2] Golrokhi et al., Applied Surface Science, 364, 789–797 (2016)
[3] Kariniemi et al., Chemistry of Materials, 23, 2901 (2011)
[4] Hauder et al., Applied Physics Letters, 78, 6 (2001)

Wide-band gap semiconductors have attracted great research attention due to their unique properties such as operation at much higher voltages and frequencies than conventional semiconductors. Our target is to engineer these materials for device applications in flexible and wearable technologies, which generally require lower deposition temperatures. Towards this goal, we have investigated the low-temperature self-limiting growth of Ga₂O₃ films on various substrates including glass and Kapton via inductively-coupled plasma-enhanced atomic layer deposition. Extended 300-cycle long runs were carried out to grow thin films on Si(100), glass, and Kapton using triethyl-gallium (TEG) as metal precursor and variants of Ar/O₂ plasma (Ar/O₂ with different partial pressures and O₂-only with different flow rates) as metal precursor and oxygen co-reactant, respectively. Growth experiments have been performed within 150 - 400 °C substrate temperature range at 300 W rf-plasma power. Ex-situ ellipsometry was employed to measure the thickness and optical properties of the films. When O₂-only plasma was used, the thickness of the films ranged between 19.32 - 16.88 nm and the as-grown refractive indices stayed fairly close between 1.73 - 1.71 within the scanned temperature range. However, addition of Ar gas to the plasma drastically reduced the refractive index of the films from 1.72 to 1.54, reminding of a possible plasma re-deposition of carbon-rich reaction byproducts or increased incorporation of oxygen. X-ray diffraction (XRD) showed that the as-grown Ga₂O₃films exhibit amorphous structure irrespective of substrate temperature in this study. Furthermore, the effect of post-growth thermal annealing is investigated on the enhancement of crystalline quality (which indicates conversion from amorphous to monoclinic β-Ga₂O₃ crystal phase) and electro-optical properties of the films. Also, X-ray photoelectron spectroscopy (XPS) measurement results of the Ga₂O₃ samples grown under varying plasma (Ar/O₂, O₂-only) will be discussed, which provides additional insight into the elemental composition of the films that might help to understand the changes in the structural, optical, and electrical properties under varying plasma chemistry. A comparative material properties analysis based on the substrate utilized will be carried out which will reveal how the film properties of as-grown Ga₂O₃ are affected by the substrate material. A future outlook will be provided to overcome the challenges to achieve device quality layers on low-temperature compatible flexible substrates.

In this presentation, we propose an alternative synthesis method to deposit thin film material hexagonal boron nitride (h-BN) assisted by microplasma source. h-BN thin films are the focus of interest for electronic, optoelectronic applications due to its wide band gap semiconductor. In other hand, microplasmas have attracted growing attention in recent years because of possible applications in other fields such as surface treatment, light sources and nanomaterial synthesis [1]. They can be generated at high pressure which is a favourable condition to better dissociate N₂ molecules, a prerequisite to the synthesis of nitride materials at lower temperatures compared to conventional processes. Our deposition reactor is composed of two chambers and the micro-plasma is located at the junction between them. The plasma source consists of an anode-dielectric-cathode sandwich through which one hole of 400 µm in diameter are drilled. The higher pressure chamber (several tens of mbar), favors the production of high density plasma, and consequently high nitrogen dissociation, while the lower pressure chamber (several mbar) limits the nitrogen recombination. The polarizable and heating substrate holder is located in the lower pressure chamber where the boron precursor (BBr₃) is injected. The polarization of the substrate holder allows the discharge to be expanded from the hole to the substrate. Suitable conditions can be defined to achieve h-BN thin film synthesis and we report the growth of h-BN on 2 inch-silicon substrates at temperatures below 1000°C. The deposited films are then characterized by XRD and Raman spectroscopy to evaluate the phase purity and quality, and by SEM and TEM to observe the surface morphology and the crystallinity of the material.

[1] T. Nozaki et al., Nanotechnol., 18, 235603(2007).

Electron enhanced atomic layer deposition (EE-ALD) can dramatically reduce the temperatures required for film growth. Temperature reduction is possible because of electron stimulated desorption (ESD) of surface species. The desorption process creates highly reactive surface sites. Precursors can then adsorb efficiently on the reactive surface sites. Our work has demonstrated the EE-ALD of GaN [1], Si [2], BN [3] and Co at room temperature.

Film growth has been performed using alternating exposures of chemical precursors and low energy electrons. In situ ellipsometry measurements have monitored linear film growth versus number of reaction cycles. Additionally, we have observed the dependence of the EE-ALD growth rates on electron energy. Maximum growth rates have varied from 0.3 Å/cycle for Si films at 100-150 eV [2] to 3.2 Å/cycle for BN films at 80-160 eV [3]. Recent measurements have also obtained growth rates of 1.0 Å/cycle for Co films at 140 eV.

Co film growth was performed using sequential cobalt tricarbonyl nitrosyl (CTN, Co(CO)₃NO) exposures and low energy electrons. The electrons desorb the CO and NO ligands from CTN on the surface and produce active sites for additional CTN adsorption. A hot filament electron flood gun was initially used as the electron source. One difficulty with the electron flood gun is the long cycle times of > 500 seconds. Much of this cycle time is consumed protecting the flood gun filament from precursor exposures and the long duration of the electron exposure due to the limited current of the electron gun. The cycle time could be reduced significantly using a more robust and higher flux electron source.

A new hollow cathode plasma electron source (HC-PES) has been developed to reduce the cycle time during EE-ALD. The HC-PES has a >600X increase in electron current compared with the electron flood gun. The HC-PES also eliminates the warm-up and cool-down time of the filament of the electron flood gun. The electron current from the HC-PES can be switched from nanoamps to milliamps in < 10 ms. The HC-PES is also chemically insensitive and reduces the need for pumping out the chamber following CTN exposures. This presentation will highlight Co EE-ALD performed using this new HC-PES.

[1] Jaclyn K. Sprenger, Andrew S. Cavanagh, Huaxing Sun, Kathryn J. Wahl, Alexana Roshko and Steven M. George, “Electron Enhanced Growth of Crystalline Gallium Nitride Thin Films at Room Temperature and 100°C Using Sequential Surface Reactions”, Chem. Mater. 28, 5282-5294 (2016).
[2] Jaclyn K. Sprenger, Huaxing Sun, Andrew S. Cavanagh and Steven M. George, “Electron-Enhanced Atomic Layer Deposition of Silicon Thin Films at Room Temperature”, J. Vac. Sci. Technol. A 36, 01A118 (2018).
[3] Jaclyn K. Sprenger, Huaxing Sun, Andrew S. Cavanagh, Alexana Roshko, Paul T. Blanchard and Steven M. George, “Electron-Enhanced Atomic Layer Deposition (EE-ALD) of Boron Nitride Thin Films at Room Temperature and 100°C”, J. Phys. Chem. C 122, 9455-9464 (2018).

GaN is the most significant binary member of the wide bandgap III-nitride material family, with a bandgap energy of 3.4 eV, and is heavily commercialized in applications such as blue light emitters for white-light LEDs, UV photodetectors, high-electron mobility transistor (HEMT)-based transistors for high-speed, high-frequency, and high-power electronics. Molecular beam epitaxy (MBE) and metal-organic chemical vapor deposition (MOCVD) are the well-known high-temperature epitaxial growth methods used for GaN deposition, whereas thermionic vacuum arc (TVA), pulsed laser deposition (PVD), reactive magnetron sputtering are among the low-temperature GaN deposition techniques. With the advent in plasma-assisted atomic layer deposition (PA-ALD) within the last decade, an increasing interest has been focused on the low-temperature self-limiting growth of III-nitride films using PA-ALD.
While a few number of groups have been active in the film growth and material characterization aspects, reported device applications have been quite limited. In this study, we aim to demonstrate that finely tuned PA-ALD processes can be utilized to synthesize crystalline GaN layers on temperature-sensitive flexible substrates, which can function as electronic device layers for thin-film transistors (TFTs). We have grown GaN thin films at temperatures less than 200°C on Si, glass, and flexible polymeric substrates using PA-ALD featuring capacitively coupled hollow-cathode stainless-steel plasma source. GaN deposition experiments were carried out using two different organometallic gallium precursors: trimethylgallium (TMGa) and triethylgallium (TEGa) to investigate optimal experimental parameters for high crystalline quality for each precursor molecule. The growth process was conducted within an in-situ ellipsometer-integrated compact PA-ALD reactor having a base pressure of 20 mtorr. RF-plasma parameters (power, exposure time, plasma gas composition) and substrate temperature were mainly studied as variable process parameters for each deposition process.
For the film thickness and optical constants (refractive index and absorption coefficient), in-situ multi-wavelength and ex-situ spectroscopic ellipsometry measurements and data fitting analysis were carried out. Synthesized GaN films turned out to be transparent non-absorbing within the visible spectrum along with refractive index values approaching 2.2 at 632 nm. The elemental composition was determined using x-ray spectroscopy (XPS). Below detection limit carbon along with relatively low oxygen concentration (< 5 at. %) was observed within the bulk of ALD-grown GaN films. For the structural characterization of the grown films, grazing-incidence x-ray diffraction (GIXRD) analysis was conducted to analyze crystalline quality and crystal grain size. XRD analysis showed single-phase hexagonal wurtzite GaN crystal structure with (002) dominant crystalline orientation. X-ray reflectivity (XRR) was used to comprehend the surface roughness and film density. High-resolution transmission electron microscopy (HR-TEM) analysis confirmed the crystalline structure of GaN films with grain sizes larger than 10 nm.
As proof-of-concept device application, we will present our metal-semiconductor-metal (MSM) photodetector characterization results which were fabricated on GaN device layers grown on glass and flexible polymeric substrates. A comparative analysis will be carried out to understand the influence of substrate material as well as the impact of mechanical stress on the flexible device performance. A future outlook will be provided to improve material and device characteristics towards implementing ALD-grown III-nitride layers for flexible/wearable device technology.

Temperature is the key parameter for controlling the phase and film morphology during chemical vapor deposition (CVD), since precursor decomposition, growth rate and obtained phases are depending on the very same. Anyhow, far from equilibrium phases and morphologies are not accessible solely by temperature but also by the application of external electric and magnetic fields. While electric fields often result in additional heating of the specimen, static magnetic (DC) fields do not lead to additional temperature gradients inside the samples and thus offer additional control on phase and morphology of the as-obtained films. In this presentation, magnetic field-assisted CVD (mfCVD) will be presented as novel and selective tool for the phase selective deposition of (magnetic) transition metal oxides and nitrides. The field-matter interplay at various temperatures and field strengths will be highlighted and consequences on physical, morphological and chemical properties of unaltered films will be discussed. Electron microscopy will demonstrate selective influence of the pristine film morphologies, while X-ray diffraction and absorption measurements will reveal changes in the obtained phase evolution and chemical surrounding. Electric, magnetic and catalytic property measurements of oxide films formed in different field environments will finally show the potential of mfCVD as additional manufacturing technique for the synthesis of functional materials.

Thin films are ubiquitous in the preparation of crystalline silicon solar cells. With the introduction of the PERC technology also the method of atomic layer deposition (ALD) has been introduced in high volume manufacturing in photovoltaics. Currently, the technique is quickly gaining market share for the deposition of ultrathin Al₂O₃ nanolayers for rear side surface passivation. The advantages of ALD are that it is scalable, that it is well suited to prepare high quality and uniform nanolayers, and that it is a "soft" deposition technique preventing interface damage. In this presentation the state of the art of ALD for silicon photovoltaics will be discussed as well as some ongoing developments and potential new applications. This includes new materials for surface passivation, nanolayers for passivating contacts, transparent conductive oxides as well as applications of ALD for polysilicon passivating contact solar cells.

One high efficiency crystalline silicon (c-Si) solar cell that has potential to be cost-competitive with Al-BSF devices while maintaining the passivation quality of PERC structures is the silicon heterojunction. Most noteably is the Heterojunction with Intrinsic Thin-layers (HIT) solar cell, which chemically passivates the surface of the c-Si with intrinsic hydrogenated amorphous silicon (a-Si:H) while allowing charge carriers to conduct through the contact. Doped a-Si:H layers then create the potential gradient necessary for carrier diffusion and charge collection to occur.

Even with it's high performance and simplified fabrication process, the HIT solar cell has potential to be improved upon. The a-Si:H layers produce parasitic absorption of high energy light. Many groups have also found the a-Si:H contacts to be sensitive to annealing treatments above 200°C. This limits the processing space for the metallization step, which typically occurs at much higher temperatures.

In this work we use Atomic-Layer-Deposited (ALD) MoOx as a hole-selective contact to c-Si in combination with thin SiOx and Al2O3 passivation layers to study the solar cell parameters of carrier selective contacts without the use of a-Si:H. While the optical properties of MoOx present a significant reduction in parasitic absorption compared to a-Si:H, the temperature stability of the material is still in question. Many groups have been unable to anneal MoOx based contacts above 120°C without degrading the electrical properties of the contact. We show that by using a thin Ni capping layer before Al metallization, the contact remains stable up to 300°C with contact resistivities below 10 mΩ-cm2. This presents significant improvements on the thermal budget of the MoOx processing sequence and will allow for more appropriate contact formation steps.

Using Ultraviolet Photoelectron Spectroscopy we measure a work function of 6.2 eV in our 5nm MoOx contact. We simulate the band-bending and hole concentration at the c-Si surface as a function of the MoOx contact work function and show that our films exhibit sufficient hole-selective properties. Additionally, High Resolution Transmission Electron Microscopy images show that when a thin Ni capping layer is not used prior to Al metallization, a 2-3 nm Al2O3 layer forms at the Al/MoOx interface. This insulating interlayer contributes to a large barrier to hole transport, making the Al/MoOx contact incompatible with high efficiency heterojunction solar cells. The Al/Ni/MoOx contact, however, exhibits no such interlayer. This suggests that Ni may act as a diffusion barrier to O species during solar cell fabrication.

Finally, we show that by using a thin (∼1 nm) Al2O3 passivation layer at the MoOx/c-Si interface, we are able to achieve a minority carrier lifetime of over 1 ms on n-type c-Si.

Recently, Yang et al. have reported 22.1% efficient c-Si solar cell by applying an atomic-layer-deposited (ALD) TiO_x thin-layer as electron contact to n-type base [1]. The origin of the electron selectivity of TiO_x has been ascribed to the asymmetric current flow at the (n) c-Si/TiO_x interface where the conduction band offset is much lower than the valence band offset. On the other hand, we recently found that TiO_x can be tuned from electron to hole selectivity by controlling the ALD condition etc. [2]. This offers an interesting possibility that TiO_x can be used as a hole selective contact alternative to the widely-used p-type a-Si:H and transition-metal oxides such as MoO_x, WO_x and V₂O_x. In this contribution, we show for the first time that TiO_x thin layers can act as efficient passivating hole selective contacts.
TiO_x thin layers were deposited by thermal-ALD on c-Si (FZ, 1 Ωcm, (100), planer, n-type). Firstly, carrier selectivity of the deposited TiO_x was studied by measuring V_oc of solar cells. To decouple the carrier selectivity from its surface passivation an intrinsic a-Si:H buffer layer was inserted between c-Si and TiO_x. A standard SHJ structure of either (i-n) a-Si:H/ITO or (i-p) a-Si:H/ITO stack was formed as a counter electron or hole contact, respectively. It is found that V_oc of the solar cell is 200-400 mV higher when using our TiO_x as hole contact than using it as electron contact. The SPV measurement showed that the TiO_x induces large band bending (~900 mV) with respect to (n) c-Si while almost no band bending is created when deposited on (p) c-Si. This implies the presence of the negative fix charge in the TiO_x, which is considered as one of the origins of the observed hole selectivity of the TiO_x. Furthermore, we found that the carrier selectivity of TiO_x depends significantly on the work function (WF) of the capping metal (or TCO) contact. The V_oc is increased monotonically with increasing the WF of the capping layer when using TiO_x as hole contacts. This indicates that band bending in (n) c-Si is significantly influenced by the WF of the capping layer, as it is well-known in the MIS contact system. By using an ITO/TiO_x/(i) a-Si:H stack as an emitter layer on (n) c-Si, we obtained a relatively high V_oc of 650 mV. Furthermore, the TiO_x is also found to act as a good passivation layer with respect to c-Si. An effective lifetime of >1 ms was obtained by depositing TiO_x on (n) c-Si without a-Si:H buffer layer. By optimizing both the surface passivation and the hole selectivity of the TiO_x layer, we attained solar cell efficiencies of >18%, demonstrating that TiO_x has potential of working as an efficient passivating hole selective contact. We discuss the origin of the hole transport in the TiO_x which contradicts to the previous transport model based on the band alignment at the TiO_x/Si interface.
[1] X. Yang et al. Prog. Photovolt: Res. Appl. 25, 896 (2017). [2] T. Matsui et al. Energy Procedia 124, 628 (2017).

The excellent electrical, optical and structural tunability of doped zinc oxide (ZnO) makes it a very good candidate for the replacement of indium-based material in the manufacturing of transparent conductive oxides. In this work, we present a comprehensive investigation of ALD grown hafnium doped ZnO within the context of its integration in different solar cells architectures. Specifically, we focus on the low range doping region, where Hf substitution is believed to be the key for band gap tunability without negatively effecting the carrier transport behavior. Scanning and transmission electron microscopy (SEM and TEM), x-ray diffraction, Kelvin probe force microscopy, Hall-effect measurements, spectrophotometry and ellipsometry were utilized to provide conclusive evidence of the suitability of Hf doped ZnO in different solar cells architectures. A band gap increase is being observed, as well as an increase in transmittance with doping. Electrically, doping is causing a decrease in resistivity and in work function. These results are interpreted in light of first-principles density functional theory simulation (DFT) to elucidate the mechanisms responsible for the electronic and electrical properties of Hf doped ZnO. DFT calculations predict a modification in the band structure of ZnO when Hf is substituted and/or embedded in the ZnO matrix as HfO₂ phases. The experimentally measured and theoretically calculated modifications in the properties of the ZnO with Hf doping, validates its compatibility with different solar cells architectures.

Antimony sulphide (Sb₂S₃) is another attractive photovoltaic material in the chalcogenide group and has drawn a great attention worldwide in the last decade. In contrast to the widely investigated and commercially competitive thin film solar cells such as CuIn_xGa_1-xSe₂ and CdTe, Sb₂S₃ is non-toxic and exists naturally as stibnite minerals, with both while Sb and S are bothas earth abundant elements . Sb₂S₃ is a binary compound with a single phase, consisting of linked one-dimensional ribbons. Such a ribbon structure provides a preferential pathway for electron transfer if withalong the desired orientation<span style="font-size:10.8333px">.</span> Antimony sulphide has high absorption coefficient of α>10⁵ and a bandgap of ~1.7eV, making it a suitable top cell candidate for tandem solar cells with silicon to overcome the single-junction Shockley-Queisser efficiency limitation. In the last decade Sb₂S₃ has been widely utilized as an efficient sensitizer in dye sensitized solar cells fabricated by chemical bath deposition . However, the CBD method is time-consuming with many undesirable oxide by-products, which usually require complex post treatment for the removal of these residuals.Moreover, reported high efficiency Sb₂S₃ solar cells have utilized unstable and expensive organic hole transport materials such as P3HT Spiro-OMeTAD, and PEDOT:PSS<span style="font-size:10.8333px">.</span> While inorganic hole transport materials such as NiO and CuSCN have been investigated, the results have not been promising. Recently, Lijian et al. used V₂O₅ as the hole transport layer (HTL) and obtained a PCE of 4.8%, demonstrating the highest efficiency of a fully inorganic planar Sb₂S₃ solar cells up to date. However, the open circuit voltage of such the record cell is 550 mV, which indicates a large Voc deficit implying there is still much work to be done to improve the Sb₂S₃ absorber quality and interface engineering. Recently, a push towards dry vacuum-based methods, such as the thermal evaporation and atom layer deposition, have been employed to grow high quality Sb₂S₃ and Sb₂Se₃ in a relatively clean environment. Additionally, the rapid thermal evaporation (RTE) method has been recognized as an effective and reliable method to grow Sb₂S₃ thin films, achieving an efficiency of 3.5% with a high Voc of 710mV when using CdS as electron transport layer (ETL)However, the reported orientation of the Sb₂S₃ is not well controlled when deposited by the RTE method. In this work, we report the first fabricated Sb₂S₃ thin films with vertical orientation by VTD method . To better understand the key factor that enables the vertical growth of Sb₂S₃, we use the RTE method as a reference, which does not create vertically aligned Sb₂S₃ crystals on a CdS buffer layer. We achieved the an efficiency of 4.73% with a high Voc of 710 mV by using the iTO/CdS/Sb₂S₃/Gold configuration via VTD method compared to 370 mV using RTE method. We propose a simple model to describe the growth process.

The field of c-Si photovoltaics has strongly diversified in recent years with the advent of a wide variety of novel passivation and passivating contact materials. Recently, we have demonstrated excellent surface passivation using stacks of ultrathin (~1.5 nm) RCA SiO_x capped with ALD ZnO/Al₂O₃, with an implied open-circuit voltage (iV_oc) of 725 mV on planar c-Si(n) wafers.[1] Within this SiO_x/ZnO/Al₂O₃ stack, the RCA SiO_x enables chemical passivation, similar as in poly-Si passivating contacts. The Al₂O₃ layer on top serves as a dense capping layer: It prevents effusion of H from the ZnO upon annealing, which is needed to hydrogenate the SiO_x.
The unique aspect of the (doped) ZnO is that it is suited as an antireflection coating (simulated J_sc of 41.6 mA/cm²) which is also conductive (< 1 mΩcm). Therefore, if a proper tunnel contact between c-Si/SiO₂/ZnO can be made, the ZnO could serve as a full-area passivating, antireflective and lateral transport layer on the front side of a c-Si solar cell. In this work, we demonstrate several crucial steps to enable this application in industrial cells by looking into the influence of texture and doping level, oxide preparation method, metal contacting and thermal stability.
Firstly, we verified that the stack also passivates on textured c-Si(n) wafers (iV_oc = 728 mV) and that its passivation on n⁺ diffused surfaces (100 Ω/sq) is on par with industrial SiN_x. The stack is thermally stable up to ~550 ^oC, which is not firing-compatible, but allows for a much higher paste curing temperature than for HIT-type cells.
Secondly, we show that the SiO_x can be prepared in many ways (RCA, LTO, UV/O₃, NAOS) which all yield good passivation. UV/O₃ however yields the best passivation and is a room-temperature, single-sided treatment which allows for accurate control over the oxide thickness (< 1.65 nm).
Thirdly, in order to be able to contact the ZnO by metal, we can selectively remove the insulating Al₂O₃ capping layer from the ZnO after hydrogenation by a wet-etch. Interestingly, no proper tunnel contact (>10 Ωcm²) can be made on 3 Ωcm c-Si(n) wafers, whereas a “first-try” value of ~0.1 Ωcm² was obtained on n⁺ diffused surfaces (100 Ω/sq). This contact resistivity is sufficiently low to use ZnO as a “hybrid” homo/heterojunction contact on n⁺ surfaces: the n⁺ doped Si surface provides electron-selectivity and facilitates tunneling, whereas the ZnO provides full-area passivation and aids in lateral transport, potentially allowing for higher Ohmic FSFs.
Ongoing work focuses on the effect of the doping levels of both the c-Si and ZnO and the integration of the ZnO on the front of a PERC-type cell.

[1] B. W. H. van de Loo, B. Macco, J. Melskens, W. Beyer, and W. M. M. Kessels, “Silicon surface passivation by transparent conductive zinc oxide,” J. Appl. Phys., vol. 125, no. 10, p. 105305, 2019.

Surface passivation of crystalline Si (c-Si) is a key enabler for achieving high efficiency c-Si solar cells. While an Al₂O₃ surface passivation layer grown by atomic layer deposition (ALD) is known for excellent surface passivation for p-type Si from the negative fixed charges in the Al₂O₃ layer, low refractive index of Al₂O₃ demands an additional anti-reflection coating to suppress optical reflection, which leads to extra capital cost in c-Si solar cell manufacturing. Therefore, the multi-functional passivation layer that can provide both the high level of passivation quality and optimum optical property is necessary for high efficiency and cost-effective silicon solar cells.
Here, we designed the multifunctional Al-doped TiO₂ passivation layer using ALD single process for low cost high efficiency silicon solar cells. TiO₂ film grown by ALD is a promising candidate for front multifunctional passivation layer of p⁺-emitter/n-base structure due to negative fixed charge densities and appropriate refractive index for anti-reflection coating of silicon solar cell. ALD process allows to control the composition accurately and provide excellent passivation quality with thin films. In this work, we controlled the Al concentration in TiO₂ film with amorphous phase from 0 to 15.5 % by adjusting the cycle ratio of Al₂O₃ to TiO₂ in ALD process. We then successfully demonstrated that the fixed charge densities of Al-doped TiO₂ passivation layer can be controlled from -8×10¹¹ cm^-2 to -3×10¹² cm^-2 by varying the amount of Al concentration. As a result, we achieved implied V_oc up to 709 mV with 15nm thick Al-doped TiO₂ on n-Si by maximizing field-effect passivation. We also investigated the effect of film thickness on surface passivation quality. By Al doping in TiO₂ passivation layer_, it showed significant enhancement of passivation performance compared to TiO₂ film from 5 to 55 nm thickness. Al-doped TiO₂ film provided high level of passivation quality leading to implied V_oc of 700 mV from 15 to 55 nm. Finally, we have conducted research to reduce the total reflectance of silicon by applying Al-doped TiO₂ passivation layer of which the refractive index became close to 2.3, ideal refractive index for silicon solar cell. Therefore, we successfully demonstrated that the ALD Al-doped TiO₂ multifunctional passivation layer is a suitable candidate for high efficient Si solar cells based on the p⁺-emitter/n-base structure with excellent optical property and outstanding passivation characteristic.

The growth of oxides by atomic layer deposition (ALD) over lead-halide perovskites in solar cells is attracting increasing attention for improving environmental and mechanical stablity. A wide range of materials have now been investigated, including SnO₂, TiO₂, Al-doped ZnO and zinc tin oxide. These oxide overlayers have led to unencapsulated perovskite solar cells achieving stable performance for 4500 h in ambient air. However, the range of growth temperatures that can be used to grow the oxide overlayers over the perovskite films is restricted to typical values of only 60 - 100 ^○C due to the low stability of the perovskites. This limits the mobility and density of the oxide films achievable. In this work, we show that we can open up the processing window of oxides grown over lead-halide perovskites by using atmospheric pressure chemical vapor deposition (AP-CVD). This technique yields oxide films with similar uniformity, density and electronic properties as ALD films at similar growth temperatures, but with orders of magnitude higher growth rates [1]. We investigate the growth of TiO_x over thermally-sensitive CH₃NH₃PbI₃ films. We achieve a growth rate of 1.19 ± 0.04 nm s^-1 at a deposition temperature of 150 ^○C, which allows 7 nm TiO_x films to be grown in 6 s (compared to >30 min for ALD). We show that this rapid deposition enables TiO_x to be directly grown on CH₃NH₃PbI₃ films without damage to the bulk or surface, as shown by our X-ray diffraction, X-ray photoemission spectroscopy and time-resolved photoluminescence measurements. Indeed, we show that the TiO_x overlayers can be grown at temperatures exceeding 180 ^○C without a significant drop in efficiency in CH₃NH₃PbI₃ solar cells. These results can be generalised to triple-cation provskite devices, as well as to AP-CVD SnO_x overlayers. In particular, we show that the conformal nature of the oxide overlayers lead to perovskite devices with improved performance (reaching 19.7% for triple-cation perovskite devices using a 60 nm SnO_x overlayer). Our work demonstrates AP-CVD to be a versatile technique for growing high-quality oxides over a wide range of processing conditions.

Reference
[1] R. L. Z. Hoye, et al., ACS Appl. Mater. Interfaces, 2015, 7, 10684

As new energy storage materials and lithium battery designs are developed to meet the ever-increasing demands for higher capacities and longer cycling and calendar lives, thin film deposition technology provides the capability of engineering key interfaces within electrochemical cells. This presentation will describe our efforts to address challenges associated with lithium metal and silicon anodes. Thin films are integrated as ultrathin solid electrolytes in Li metal batteries and as protective coatings on Si anodes to limit undesirable side reactions.

Key requirements for electrolytes in solid-state lithium metal batteries are a large electrochemical stability window and low area-specific resistance (ASR). Solid electrolytes must also possess robust mechanical properties to accommodate large-scale production and integration into conventional lithium battery cell designs. To realize these properties, 50 nm-thick films of lithium phosphate oxynitride (Lipon) were deposited onto microporous polymer separators (Celgard) using RF magnetron sputtering. These separators provide a low ASR due to the thin, dense Lipon film; the total resistance of the separator was determined to be 40 Ω cm² in alkyl carbonate electrolytes, which is much lower than traditional ceramic electrolyte membranes, such as those fabricated from Garnet and NASICON-class of solid electrolytes. Furthermore, these composite separators inhibit chemical cross-diffusion and reaction between anode and cathode in both Li-S and Li-LiMn₂O₄ cells. Both the performance of these hybrid separators in practical lithium-metal batteries and fundamental studies of ion transport at the Lipon-liquid electrolyte interface will be presented.

Silicon is also an intriguing next-generation anode offering charge capacities comparable to lithium metal, yet significant challenges arise from the >300% volume expansion of Si during lithiation. To address continual electrochemical reduction of lithium ion battery electrolyte on Si anodes, nanoscale, conformal polymer films were synthesized as artificial solid electrolyte interface (SEI) layers. Initiated chemical vapor deposition (iCVD) was employed to deposit poly(1,3,5,7-tetravinyl-1,3,5,7-tetramethylcyclotetrasiloxane) (pV4D4) onto silicon thin film electrodes. 25 nm-thick pV4D4 films on Si electrodes improved initial coulombic efficiency by 12.9% and capacity retention over 100 cycles by 64.9% relative to untreated electrodes. PV4D4 coatings also improved rate capabilities, enabling higher lithiation capacity at all current densities. Post-cycling FTIR and XPS showed that pV4D4 inhibited electrolyte reduction and altered the SEI composition, with LiF formation being favored. This work will guide further development of polymeric artificial SEIs to mitigate electrolyte reduction and enhance capacity retention in Si electrodes.

In this work Chemical vapor deposition (CVD) and Atomic Layer Deposition (ALD) has been combined to elaborate electrodes for MicroSuperCapacitors (µSCs) devices. A great deal of attention has been focused on µSCs, for which large series of nanostructured active materials have been developed.
We have demonstrated through comprehensive investigations the interest of doped silicon nanostructures grown by CVD with Vapor Liquid Solid (VLS) method as electrodes materials for µSCs using ionic liquid electrolytes [2,3]. The fine morphological tuning of the nanostructure allowed by the bottom-up approach enables specific designs of electrode architectures, with a considerable leeway compared to other techniques. Such latitude allows optimizing porosity and ionic and electronic pathways while keeping robust mechanical and thermal performances, depending on the target application. Si Nanostructures (SiNs) such as Si nanowires and Si nanotrees have displayed excellent electrochemical performances being stable over than 1 million cycles of galvanostatic charge/discharge under a 4 V wide electrochemical windows in EMI-TFSI ionic liquid, with large power densities of 10 mW.cm^-2 and good capacitance values of 0.5 mF.cm^-2 at high current density of 0.5 mA.cm^-2 [4].
However a major silicon weakness which was still hindering its use with aqueous electrolytes is the native uncontrolled growth of silica when subjected to ambient atmosphere. Here we have developed a highly conformal passivation coating of a nanometric high-k dielectric layer of Al₂O₃ based on the rising ALD technique. ALD has proven to allow a nanometric thickness control of the deposited layer while being highly conformal and covering. Electrochemical stability performances in ionic liquid, were enhanced allowing symmetric 2 electrodes devices to reach an unprecedented cell voltage of 5.5 V [5], improving energy and maximum power densities compared to pristine nanostructured silicon. The cyclability was also largely enhanced, with only 3% capacitance fade after 10⁶ galvanostatic charge/discharge cycles at 4 V, and no degradation even after several 10⁵ resilience cycles over a 5 V window [5]. Moreover, the protective alumina layer enables the use of aqueous electrolytes for nanostructured Si based µSCs, which significantly increases the specific power of the devices up to 200 mW.cm^-2 at 0.5 mA.cm^-2 while keeping the capacitance performances at 0.5 mF.cm^-2. Furthermore the system is remarkably able to retain 99% of its initial capacitance after 2 billion galvanostatic charge-discharge cycles at high current density of 0.5 mA.cm^-2 in an aqueous electrolyte of Na₂SO₄.
Eventually we have realized a composite SiNs-Alumina-conducting polymer as electrodes for pseudocapacitive device. This device exhibits promising performances with a specific energy of 2 Wh.kg^-1 and a power density of 300 W.kg^-1 at a current density of 1 A.g^-1. The µSCs was able to retain 80% its initial capacitance after 50,000 galvanostatic charge-discharge cycles at 0.5 A.g^-1 [6].
[1] Beidaghi M., Gogotsi Y., Energy & Environ. Sci. 2014, 7 (3), 867-884
[2] Thissandier F., Gentile P. ; Pauc N., Brousse T. , Bidan G. , Sadki S., Nano Energy 2014, 5, 20-27
[3] Thissandier F., Gentile P., Sadki S., 2014, Journal of Power Sources 269, 740-746
[4] Gaboriau D., Aradilla D., Gentile P., Sadki S., RSC Advances, 2016, 6, 81017-81027
[5] Gaboriau D., Boniface M., Valero A., Aldakov D., Brousse T., Gentile P., and Sadki S., ACS Appl. Mater. Interfaces 2017, 9, 13761−13769
[6] Valero A., Mery A., Gaboriau D., Gentile P., and Sadki S., ACS Appl. Energy Mater. 2019, 2, 436−447

ZnO has been a well-explored oxide thermoelectric material owing to its low toxicity, high abundance and good chemical stability. Among the deposition techniques used for ZnO thin film fabrication, atomic layer deposition (ALD) is recently gaining attention due to its excellent film uniformity and precise thickness control down to the Angstrom level. It is also an ideal technique for depositing doped ZnO for thermoelectric applications because optimum electrical properties can be achieved by actively controlling the dopant levels. However, reports on the thermoelectric properties of doped ZnO fabricated by ALD have been limited to a few dopants such as Al and Ga [1-2]. Herein, the thermoelectric properties of ALD Hf-doped ZnO (HZO) thin films are reported for the first time. On Si and glass substrates, ALD growth was performed by sequential pulsing of diethylzinc (DeZn) and H₂O to form the ZnO, and gas pulsing of tetrakis(dimethylamido)hafnium(IV) (TDMAHf) and O₂ to incorporate the dopants. Three supercycles of ZnO and dopant were deposited, wherein one supercycle is composed of varying nominal dopant amounts (0%, 1%, 2%, 3% and 4%). The deposition temperature was 443 K and nitrogen was used as the carrier gas. Thermoelectric properties were measured from 303 to 453 K (heating) and back (cooling) to observe hysteresis. Below 453 K, pure ZnO sample exhibited higher σ compared to all the HZO samples. However, at 453 K, an increase in electrical conductivity of HZO thin films was observed when a minimum of 3% Hf was added. A maximum of 94.9 S/cm was reached for 4% HZO, which was higher than in pure ZnO. In addition, a large hysteresis was observed with the pure ZnO sample during the cooling scan, leading to a sixfold decrease in electrical conductivity. The Seebeck coefficients are almost similar for the ZnO and HZO samples throughout all temperatures during the heating scan, but a large hysteresis was likewise observed during cooling for the pure ZnO and the lightly doped (1% and 2%) HZO. The 3% and 4% HZO samples displayed minimal hysteresis, leading to a highly stable power factor throughout both heating and cooling stages. A maximum power factor of 157 μW/mK² was obtained for the 3% HZO sample at 453 K, exceeding that of pure ZnO. Doping with 3% Hf therefore leads to an enhanced thermoelectric power factor as well as improved stability compared to pure ZnO thin films deposited by ALD.

[1] M. Ruoho, V. Pale, M. Erdmanis and I. Tittonen, Appl. Phys. Lett. 2013, 103, 203903.
[2] S. Lee, J. Lee, S. Choi, and J. Park, Ceramics International, 2017, 43, 7784-7788.

Redox-based resistive random access memory (ReRAM) devices are intensively investigated as artificial synapses in neuromorphic applications due to their simple fabrication, switching performance and energy efficiency. For the realization of high-density synaptic networks with large connectivity two concepts are pursued. In the stacking of two-dimensional (2D) passive crossbar arrays, the number of stacked layers is limited by patterning issues. Therefore, blocks with even larger numbers of integrated synapses require real 3D stacking of the memory cells. At this end, atomic layer deposition (ALD) becomes the unique deposition method of choice. Due to its surface reaction controlled characteristic, ALD enables a conformal and pinhole-free coverage of the structured electrodes by ultra-thin metal oxide layers with high structural uniformity. The performance of nanoscale devices from ALD grown oxide thin films is thus a key factor for future 3D ReRAM architectures.
Today’s nano-sized ReRAM devices contain resistive switching layers from metal oxide films of a few nanometers in thickness. The precise control of the layer thickness directly affects the electroforming voltage necessary for the onset of the resistive switching, and also the switching reliability. In addition, the layer’s microstructure and the stoichiometry have to be controlled. Therefore, the device performance strongly depends on the precise control of the film growth behavior.
This talk focuses on the comparison of thin film properties obtained for the most studied oxides in ReRAM research, namely, TiO_x, Al₂O₃, TaO_x and HfO_x, fabricated from two different ALD processes, this is, thermal ALD using water vapor and plasma-enhanced ALD utilizing a remote oxygen plasma as the oxygen source. Conformal reproducible growth of a few nanometer-thick oxide films onto structured bottom electrodes is the challenge. Characterization combines structural film analysis with electrical device characterization providing feedback for the ALD parameter optimization. By this means reliable 2D micro- and nanoscale ReRAM devices were fabricated. The oxide layer(s) are stacked between a Schottky electrode and an ohmic electrode. Engineering of both interfaces is carried out to further improve the cell reliability. Especially, for the TiO_x-based devices it turned out to be advantageous to add a thin Al₂O₃ tunneling barrier between the Pt Schottky electrode and the oxide layer to gain the control over leakage. Further, it has been shown that the insertion of a thin TiO_x layer between the Ti ohmic electrode and the HfO_x switching layer reduces the variability in the resistance values and the set voltage. Here, the character of different oxides and the effect of stack order from combination of several oxide films is discussed, aiming at an interface design and a tuning of the switching behavior enabling a more rational design of ReRAM devices.

Resistance switching in metal-insulator-metal structures has been extensively studied in recent years due to potential applications in non-volatile memory technology as well as alternative computational frameworks^1,2. Despite their scalability (10 nm), device density (4F²), high switching speed and endurance, resistive memories still suffer from considerable performance trade-offs between low voltage operation, high on/off ratio and device variability. In this work we report for the first time sequential infiltration synthesis (SIS) as a pathway to develop ultra nanoporous oxide thin films for high performance conductive-bridge resistive memory. SIS is a recently developed modified atomic layer deposition (ALD) process whereby the inorganic precursor molecules for oxide growth infiltrate a polymer thin film matrix.^3,4 In this work we grow nanoporous alumina at 95°C, using Poly (methyl methacrylate) (PMMA) as the polymer phase and trimethyl aluminium (TMA) and water as the ALD precursors. Nanoporous alumina with 5 nm pore size and ~ 70% porosity is obtained after removing the polymer phase by rapid thermal annealing for 20 minutes in a mixture of O₂/Ar/N₂. Crossbar resistive memory devices of size ranging from 200x200 nm² to 1 µm² are fabricated with Pt as the bottom electrode, nanoporous alumina as the switching layer and Ag as top electrode. The nanoporous alumina devices demonstrate ultra-low operating voltages (~±550 mV) and ultra-high on/off ratio (> 10⁹) in addition to pulse endurance up to 1 million cycles and high temperature (125 °C) retention for up to 10⁴ seconds. The combination of ultra-low operating voltages and extremely high on/off ratio shows the best performance reported so far. In addition, the devices also exhibit low intrinsic variability with standard errors ~2% for the operating voltages, <2% for the on-state resistance and <10% for the off state resistance values. Combining high resolution transmission electron microscopy, 3D tomography and electrical characterizations we argue that the presence of high internal free surface area results in ultra-low power operation.
REFERENCES:
1. A. Chen et al., Non-volatile resistive switching for advanced memory application, IEDM Tech. Dig., 2005, pp. 746-749.
2. J. J. Yang et al., Memristive devices for computing, Nat. Nanotech., vol. 8, pp. 13-24, 2013.
3. M. Biswas et al., New Insight into the Mechanism of Sequential Infiltration Synthesis from Infrared Spectroscopy, Chem. Mat., vol. 26, pp. 6135-6141, 2014.
4. D. Berman et al., Sequential Infiltration Synthesis for the Design of Low Refractive Index Surface Coatings with Controllable Thickness, ACS Nano, vol. 11, pp. 2521-2530, 2017.

Water photosplitting is a promising way to transform the sunlight into a storable and transportable energy source. Among numerous semiconductors, Si can be used as photoanode because it absorbs in the visible range and its electronic structure is suitable to drive water photooxidation [1]. Though, Si suffers from a strong corrosion at high pH and it exhibits a high reflectivity. The present approach consists of depositing oxide and metal thin films to protect the surface and to enhance the Si electroactivity. To further improve the photoelectrode performances, various surface structuring methods are used to enlarge the light absorption, to increase the active area and to improve the charge collection.

Since Atomic Layer Deposition (ALD) is a well-adapted technique to coat both planar and tortuous surfaces, it has been used to grow a thin protective layer of TiO₂ on Si. An exhaustive investigation of the influence of the deposition parameters onto the final photoelectrochemical properties has revealed the crucial impact of the nature of the precursor (titanium isopropoxide,TTIP or tetrakisdimethylamidotitanium ,TDMAT) as well as the effect of the process temperature. A perfect stability is reached when the TiO₂ layer is grown using TDMAT at 150°C. This has been ascribed to the easier Ti–N bond breaking during the TDMAT pulse. A detailed electrochemical study has also shown that Si oxidation and etching can proceed through the TiO₂ thin film at open circuit potential (ocp) under low illumination.

To increase the photocurrent and to lower the overvoltage, a co-catalyst must be added. Metallic Ni has been deposited by PVD or electrodeposition [2,3] but the quality of the films is not fully satisfying on tortuous substrates. In the present work, Ni has been deposited onto the TiO₂-covered Si according to a two-step process: (i) conformal ALD of NiO from Ni(CpEt)₂ and O₃ and (ii) the reduction to Ni by annealing under H₂ atmosphere. The efficiency of such multilayered photoelectrodes is drastically improved using significantly less Ni and the photoanodes exhibit a long term stability.

This approach has been extended to structured Si surfaces such as macroporous [4], nanospikes [5] and micropillars [6]. The use of those simple electrochemical structuring methods has led to enhanced water photooxidation on oxide-covered Si. It has been confirmed using another protective layer: Fe₂O₃. Iron oxide is grown by ALD using Fe(CptBu)₂ and O₃. In this later case, the photocurrent is increased by a factor 20 when using Si nanospikes as substrates.

This works shows clearly how various, well-controlled, ALD processes and surface structuring methods can be combined to optimize the photoanode properties. The electrochemical investigations give new insight on the photoelectrochemical phenomena occurring at the solid/liquid junction in both operating conditions and at ocp.

[1] K. Sun, S. Shen, Y. Liang, P. E. Burrows, S. S. Mao, D. Wang, Chem. Rev. 114, 8662 (2014).
[2] S. Hu, M. R. Shaner, J. A. Beardslee, M. Lichterman, B. S. Brunschwig, N. S. Lewis, Science 34,1005 (2014).
[3] G. Loget, B. Fabre, S. Fryars, C. Mériadec, S. Ababou-Girard, ACS Energy Lett. 2, 569 (2017).
[4] L. Santinacci, M. W. Diouf, M. K. S. Barr, B. Fabre, L. Joanny, F. Gouttefangeas, G. Loget, ACS Appl. Mater. Interfaces, 8, 24810 (2016).
[5] G. Loget, A. Vacher, B. Fabre, F. Gouttefangeas, L. Joanny, V. Dorcet, Materials Chemistry Frontiers 1, 1881 (2017).
[6] F. J. Harding, S. Surdo, B. Delalat, C. Cozzi, R. Elnathan, S. Gronthos, N. H. Voelcker, G. Barillaro, ACS Appl. Mater. Interfaces, 8, 29197 (2016)

In the past decade, inorganic semiconductor light-emitting diodes (LEDs) based on the gallium nitride (GaN) material system were established in many everyday devices as an integral part of modern communication and consumer electronics. The conventional GaN-based LED technology relies on p-n junctions with embedded indium gallium nitride (InGaN) quantum wells which are used to obtain emission in the blue spectral range. However, the p-doping of GaN involves challenges leading to low conductivity of the p-GaN of several S/cm at high defect densities. Today, the technological difficulties of p-doping have not yet been completely overcome and gain new relevance by the development of three-dimensional (3D) LED structures to further increase the efficiency of LEDs.
In this work, we replaced p-GaN of conventional GaN-LEDs with a p-conductive poly(3,4-ethylenedioxythiophene) (PEDOT) by using oxidative chemical vapor deposition (oCVD) to obtain novel hybrid-LED architectures based on inorganic and organic materials. In comparison to classical wet-deposition techniques, e.g. spin-coating of PEDOT:PSS, we observed conformal, pin-hole free, ultrathin (<100 nm) coatings of oCVD-PEDOT on planar and nano-rod GaN-LED surfaces by using gaseous monomer (EDOT) and oxidant (FeCl₃). We observed stable, blue (~450 nm) electroluminescence (EL) for an applied voltage of > 3 V under ambient conditions for planar hybrid-LED structures which are directly fabricated on full 2"-wafers. We determined the current-voltage profiles of the hybrid-LEDs and obtained diode characteristic behavior. In comparison to conventional p-GaN LEDs the hybrid-LED configuration provides a more intensive, areal light emission which is attributed to the improved lateral conductivity of oCVD-PEDOT with ’face-on’ orientation of the polymer backbone. Our results demonstrate the feasibility of oCVD for producing and integrating a highly p-conductive transparent polymer thin film into planar LED devices.

Membrane-based organic solvent separations promise a low-energy alternative to traditional thermal separations but require advanced materials that operate reliably in chemically aggressive environments. While inorganic membranes can withstand demanding conditions, they are costly and difficult to scale. Polymeric membranes, such as polymers of intrinsic microporosity, are easily manufacturedinto form factors consistent with large-scale separations (e.g., hollow fibers), but perform poorly in aggressive solvents. Here, a new post-fabrication membrane modification technique, vapor phase infiltration (VPI) is reported that infuses polymer of intrinsic microporosity 1 (PIM-1) with inorganic constituents to improve stability while generally maintaining the polymer’s macroscale form factor and microporous internal structure. The atomic-scale metal oxide networks within these hybrid membranes protect PIM-1 from swelling or dissolving in organic solvents including: tetrahydrofuran, dichloromethane, andchloroform. This atomic-scale metal oxide network further decreases the molecular weight cutoff (MWCO; the smallest molecular weight the membrane “successfully” rejects) in n-heptane and toluene from a MWCO of about 600 g/mol for pristine PIM-1 thin film composite membranes to 204 g/mol for hybrid AlOx/PIM-1 membranes. The hybrid membranes further retain this MWCO and high levels of rejection (>95%) in solvents that traditionally swell or even dissolve pristine PIM-1 (such as ethanol and tetrahydrofuran). The decrease in MWCO and increase in stability of AlOx/PIM-1 hybrid membranes allows them to perform separations not only between solutes and solvents, but also separations of more challenging systems such as those comprising multiple solvents. For example, the hybrid AlOx/PIM-1 membranes are capable of enriching the toluene concentration in a mixture of 90 wt% toluene, 5 wt% 1,3,5-triisopropylbenzene, and 5 wt% 1,3-diisopropylbenzene from 90.0 wt% to 97.8 ± 0.3 wt% toluene. In this talk, we will discuss the chemical mechanisms of the infiltration process that we believe create the hybrid structures necessary to support this enhanced stability and separation performance.

Infiltration synthesis, an emerging material hybridization technique derived from atomic layer deposition (ALD), utilizes the infiltration of vapor-phase organometallic precursors into polymeric templates to generate organic–inorganic hybrids with uniquely enhanced material properties. A subsequent removal of organic matrix from the hybrid further generates functional inorganic nanostructures directly converted from the starting polymeric template, providing a new inorganic nanopatterning methodology. Fundamentally, the infiltration synthesis is mediated by the binding reaction between infiltrating Lewis-acidic organometallic precursors and Lewis-basic functional groups available in the polymer matrix. Hinging on this principle, self- assembled block copolymer (BCP) thin films (e.g., poly(styrene-block-methylmethacrylate) (PS-b-PMMA)) enable an area-selective infiltration of inorganic components owing to the intrinsically available, spatial chemical contrast (i.e., chemically reactive domains being spatially separated from non-reactive domains). While the infiltration synthesis of nanopatterned AlO_x is routinely performed by infiltrating trimethylaluminum (TMA) into BCP templates, other more functional metal oxides, such as ZnO, are not readily amenable to the nanopatterning by infiltration synthesis unless a pre-infiltration of insulating AlO_x (i.e., “AlO_x-priming”) is applied, due to relatively weaker binding affinities between precursors (e.g., diethylzinc (DEZ) for ZnO) and active polymer templates. In this work, we report the infiltration synthesis of optoelectrically active, three-dimensional (3D) pristine ZnO nanomesh architectures by combining a modified infiltration synthesis protocol called micro-dose process with hierarchically stacked self-assembled polystyrene-block-poly(2-vinylpyridine) (PS-b-P2VP) BCP thin films. The combination of new infiltration protocol and more reactive BCP template with 3D hierarchical self-assembly not only obviates the need of AlO_x-priming but also generates a new 3D ZnO nanomesh structure, which exhibits optoelectrical functionality, featuring stack-layer-number-dependent electrical conductance resembling the percolative transport originating from the intrinsic morphological network connectivity of the lamellar BCP pattern with symmetric block ratio. The results not only illustrate the first demonstration of electrical functionality based on the ZnO nanoarchitecture directly generated by the infiltration synthesis in self-assembled BCP thin films but also present a new, large-area scalable, metal oxide thin film nanoarchitecture fabrication method utilizing industry-compatible polymer solution coating and ALD. Given the large surface area, 3D porosity, and readily scalable fabrication procedures, the generated ZnO nanomesh promises potential applications as an efficient active medium in chemical and optical sensors.

Atomic layer processing is an umbrella term for several processing techniques that base on atomic layer deposition (ALD). Those include thin film coatings with inorganic or organic materials, atomic layer etching and polymer infiltration strategies.
The ALD process can be seen from various perspectives. On the one hand, it allows controlled deposition of thin films on a variety of substrates and in this way enables a modification of a given functionality of a surface or even introduction of a new functionality. On the other hand, it may be seen as a chemical reactor that allows precise dosing of a chemical, allowing for chemical interaction and modification of the substrate. Considering both points of view, the process opens large variation possibilities for a design of novel functional materials for emerging applications and devices. Among those functional materials hybrid materials play an increasingly important role. Hybrid materials are in most cases blends of inorganic and organic materials and are considered to be key for the next generation of materials research. The main goal while fabricating such materials is to bridge the worlds of polymers and ceramics, ideally uniting the most desirable properties within a singular material. Furthermore, in a well performing hybrid material the individual components will add value to their counterpart in a synergistic way.
In this talk, some approaches will be discussed that show great promise for establishing ALD as the method-of-choice for innovation in technological fields beyond the microelectronics industry. Rather than growing thin conformal films, the ALD process technology is applied to controllably grow nanoparticles on functional substrates adding value to their chemical or electrochemical properties. In an adapted processing mode, the ALD processing technology also allows infusing metals into polymeric substrates, which leads to novel material blends that cannot easily be obtained in other ways. In either of those cases the chemical or physical properties of the initial substrate are improved or new functionalities added. With some showcases, this talk will discuss approaches towards non-traditional application of ALD to fabricate novel materials with great promise in energy storage, catalysis, personal protection or flexible electronics. We will show examples where mechanical and electronic properties of polymeric materials have been significantly improved through nanoscale coatings and infiltration. Those polymers include natural polymers such as spider silk, collagen, etc. as well as technical polymers such as Kevlar, polyaniline and polythiophene. The characterization of the resulting material has been done by means of various spectroscopies, microscopies and modelling by density functional theory.

ALD-ZnO thin films can achieve conductivities of ≤3 mOhm cm without any intentional doping such as by supercycling ALD-Al2O3 to deposit Al-doped ZnO (AZO). Whereas the latter allows for conductivities down to the 10^-4 Ohm cm range, the mobility of ALD-AZO is significantly reduced by impurity scattering and the free carrier absorption is increased, which induces e.g. optical losses when applied to solar cells. Generally, two different models for the n-type conductivity in ZnO are discussed in the literature: intrinsic ZnO point defects and hydrogen. However, ZnO deposited or grown by methods other than thermal ALD rarely reach the very high conductivity range that is essential for its application as transparent conductive oxide (TCO). Hence, ALD appears to introduce a very efficient intrinsic doping mechanism into the material unlike the other methods. We will show via T-dependent Hall-measurements that the intrinsic doping is so high that it induces a semiconductor-to-metal-transition (Mott transition). By means of isotope studies (deuterium vs. hydrogen) the origin of the ~1 at% H-concentration in the ALD-ZnO thin films is determined. Using a comprehensive range of structural-chemical characterization methods (FTIR, NRA, NMR, c-AFM, XANES at the O and Zn K-edges) the bonding configuration of H in ZnO is studied in detail. Finally, different models for the incorporation of H into the ZnO network and their resulting density of states (DOS) are simulated by density functional theory (DFT) to explain the very high conductivity of ALD-ZnO.

Recently, artificial neural networks (ANN) have revolutionized the field of artificial intelligence with human-like performance in fields such as computer vision and speech recognition. However, the current computers with von Neuman -architecture are ill-suited for the massively parallel operations required for ANNs resulting in high energy consumption. Neuromorphic computing aims to greatly improve the energy efficiency by emulating the functionality and interconnectivity of the biological neural networks on the hardware level. In biological neural networks, communication between neurons is facilitated by synapses that enable combined memory and processing. Development of brain-like computers requires materials and devices that have the same functionalities as the real synapses.

Metal-organic hybrid materials (coordination polymers) present an interesting candidate for this task. Their electronic conductivity can be modulated electrochemically by controlling the oxidation state of the material, and can be used to model the synaptic weight. The interplay of the metal and organic constituents enables the tuning of the electric/electrochemical properties towards low device conductance and low power consumption. Additionally, as the redox reactions are coupled with the motion of an associated counter ion, the difference in the electron and ion mobilities could potentially be harnessed to simulate various time dependent synaptic functionalities such as short- and long-term memory. However, currently the poor processability of coordination polymers limits their applicability for nanotechnology applications.

Here we demonstrate that Molecular Layer Deposition (MLD) is extremely well suited for this application. As an extension to ALD, MLD expands the self-limiting gas-surface reactions to hybrid organic-inorganic materials with the implementation of volatile organic molecules as the co-reactant. As a solvent-free, gas-phase processing technique it is in principle also compatible with the existing CMOS processing methods. As a proof-of-concept, we developed a novel MLD process for Cu-dithiooxamidate based thin films. The growth and composition of the films is characterized with spectroscopic ellipsometry, XRR, XRD, XPS, AFM and FTIR. The resistive switching properties are explored in simple two-terminal devices arranged in cross-bar arrays of variable node sizes. The deposition process is shown to comply with the principles of ALD with surface saturation limited growth. Excellent thickness uniformity and low surface roughness (RMS < 0.3 nm) can be realized over large area substrates. The solid-state devices demonstrate an oxidation state dependent dynamic conductance range spanning over one order of magnitude. In pulsed operation mode similar to biological neurons, the devices demonstrate many of the basic synaptic functionalities. The rate of change in the device conductance is dependent on the excitation frequency and amplitude, thereby emulating spike-rate and spike timing dependent plasticity (SRDP and STDP). Furthermore, both volatile and non-volatile changes in device conductance can be realized to mimic the short and long-term plasticity (STP and LTP) of the biological synapse.

We propose and demonstrate a novel thin film deposition technique by transferring the principles of atomic layer deposition (ALD), known with gaseous precursors, towards precursors dissolved in a liquid. The technique can also be considered as a generalization of already established methods such as the ‘layer by layer’ growth or the ‘successive ion layer adsorption and reaction’ (SILAR). 'Solution ALD' (sALD) shares the fundamental properties of standard ‘gas ALD’ (gALD), specially the self-limiting growth and the ability to coat conformally deep pores. It has been already shown that it is possible to transfer standard reactions from gALD to sALD such as TiO₂ deposition. However, sALD also offers novel opportunities such as overcoming the need for volatile and thermally robust precursors.
In the following the targeted field is photovoltaic. A great deal of interest has appeared on a new generation of material for solar cells application. Among them the perovskites are particularly interesting and the most studied one is CH₃NH₃PbI₃ (MAPI). The existing deposition methods such as spin coating or vapor-deposition techniques do not allow a control at the atomic level. ALD has been used to deposit PbS but it needed a two-step conversion method to obtain a perovskite. Therefore, a new process based on sALD is developed to deposit PbI₂. It allows the use of inexpensive lead salt and it is easy to process. Then, the PbI₂ can be easily converted to MAPI. The PbI₂ deposition was achieved with Pb(NO₃)₂ and LiI via s-ALD on large samples (up to 10 cm*10 cm). The ALD behavior has been shown. The influence of the deposition parameters on the morphology, the crystalline structure and the chemical composition has been investigated by scanning electron microscopy, atomic force microscopy, grazing incidence x-ray diffraction and x-ray photoelectron spectroscopy. The growth of PbI₂ has been studied on different oxides with various chemical pre-treatments. Later, the obtained PbI₂ has been converted to MAPI by vapor annealing. The influence of the temperature, the time and the type of gas on the conversion has been investigated the different substrates. Furthermore, the influence of the ALD parameters on the properties of the perovskite layer has been studied. Thus, the perovskite layer has been integrated into a solar cell.

We report on the development of lithium compounds by atomic layer depostion and atomic layer etching for space applications related to far ultraviolet telescope mirrors and energy storage devices. Lithium fluoride is a promising next generation mirror coating material for astrophysics instruments and initial ALD prototype mirrors have exceeded the performance and stability of legacy PVD systems. We have also investigated improvements in stability obtained by the use of nanolaminates with ALD aluminum fluoride. Our HF-based approach to the deposition of these materials has also led to thermal atomic layer etching methods that can be implemented on the same aluminum mirror systems to gently remove native oxide from the aluminum surface. Exploring an interaction between trimethylaluminum and HF has been shown to greatly enhance the etch effect on aluminum oxide as a function of substrate temperature. Planned flight demonstrations of these coatings on NASA sounding rocket and CubeSat missions will be discussed.

The same material system is also a promising battery interface material, and this talk will present initial results on the passivation of Li-metal anodes using this ALD approach. The use of AlF3, LiF, and intermediate phases of the two have been investigated for their ability to improve cycling stability and capacity in test cells. The same interactions with alkali materials observed in etching experiments is shown to influence the behavior of atomic layer processing approaches on Li-metal anodes.

Highly-porous ceramics are employed in a variety of technological applications, such as filtration, catalysis, heat insulation and photonics. Photonic materials for high-temperature applications are expected to perform while being exposed to temperatures higher than at least 800 °C. Nonetheless, this exposure usually causes detrimental morphological changes with associated impairment of the photonic properties. Such properties are defined during the fabrication process and are related not only to the material composition but also to the structure present at different scales. 3D inverse opal photonic crystals (PhCs) are usually produced by the infiltration of a polymeric template either by sol-gel, colloidal routes, chemical vapor deposition or atomic layer deposition (ALD). Hereby, we demonstrate the production of ceramic-based photonic materials by combining the controlled self-assembly of monodisperse particles with ALD. A variety of ceramic PhCs produced via ALD will be presented and their behavior when exposed to high temperatures depicted. By specifically designing and controlling the ALD process, materials with tailored composition are obtained which present a photonic stopgap even after exposure at 1400 °C, a remarkable achievement concerning temperature stability. At last, an example of additive manufacturing combined with colloidal assembly (AMSA) will be presented as a demonstration of large area templating, which in combination with ALD creates new possibilities for photonic and highly-porous materials production.

ZnO nanotubes with a wall thickness on the order of nanometers were produced by atomic layer deposition (ALD) with an organic template, using diethyl zinc as Zn precursors and water as O precursors. The organic template was removed by an O₂ plasma process and a thermal annealing process, with the wall structure of the ZnO nanotubes maintained during the removal of the template. We investigated the response and selectivity of a ZnO nanotube to ethanol and hydrogen gases as a function of its wall thicknesses. The ZnO nanotubes with a wall thicknesses of 10 nm showed the highest gas response (defined as the ratio of electrical resistances, R_air/R_gas, under air atmosphere to the mixed air atmosphere with ethanol vapor) of 1184 toward 100 ppm at 450°C. The ZnO nanotube with a wall thickness of 10 nm was wholly depleted and thus highly responds to the adsorption of chemical species on its surface. Accurate control of the wall thickness of the ZnO nanotubes is crucial to make ZnO nanotube sensors with high response. The memristor is considered as the one of the promising candidates for next generation gas sensor, neuromorphic computing device. Metal-chalcogenide memristors with small thicknesses and high ionic mobility were fabricated with ALD, shadow mask patterning technique, and metal elctrode deposition.

As dimensions have been shrunk close to the limits of what is physically possible, novel materials as well as alternative device concepts and architectures are increasingly required. Examples of such materials can be logic gate spacers with lower dielectric constants, alternative channel materials with a higher mobility or metals with decreasing resistivity to name a few. These are being added to the widening list of novel materials with every new technology node. In addition and independent of the application, there is an ongoing trend of 3D integration: logic (FinFET), DRAM (trench or pillar capacitors) or NAND Flash (staircase 3D NAND) are examples of the how devices have been innovated into the third dimension to enable a continued area scaling and related performance benefit. Such 3D architectures imply the need for conformal deposition techniques such as atomic layer deposition. In addition, to reduce the level of edge placement error, self-aligned process schemes and the use of selective deposition is being explored.
In this presentation, we will discuss how the semiconductor landscape is evolving and how material innovation is an absolute must. We will highlight some of the material options that are being considered and touch upon selective deposition to facilitate integration schemes.

Together with the development of flexible and wearable devices, there is a growing demand of flexible energy storage devices as power sources. Supercapacitors, bridging the gap between batteries and capacitors, are very promising among the candidate power solutions. However, one of the key challenges for wide application of such flexible supercapacitors is that the maximum current density allowed for them is not high enough. The issue is partially induced by the structure of the electrode materials. Traditionally, randomly aligned carbon nanotubes are used as electrode material for supercapacitors, hindering the ion transport inside the electrode and therefore limiting the current density it can achieve. In this work, we demonstrate that horizontally aligned carbon nanotubes (CNTs) exihibits superior ion transport inside the electrodes, and therefore significantly enhance the maximum current density. At the same time, conjugated polymers with pseudocapacitive characteristics have drawn much attention for the application of supercapacitors to enhance their capacitance. We for the first time, demonstrate oxidative chemical vapor deposition (oCVD) synthesis of a conjugated polymer, poly(3-methylthiophene) (P3MT). With only one methyl as the side group of the thiophene ring, P3MT demonstrates higher density of effective conjugation (polythiophene structure) than other modified polythiophene-like polymers, such as PEDOT and poly(3-hexylthiophene). In this way, the lighter repeating unit increases specific capacitance. With oCVD, we also demonstrated conformal coating of P3MT on CNTs with an average thickness of ~5 nm. The resulting device exihibits high areal capacitance of 3.1 F cm^-2 at 5 mA cm^-2, with areal capacitance shown to remain high at 1.8 F cm^-2 even at a current density of 200 mA cm^-2, which is much improved compared to the state-of-the-art devices. Furthermore, little change in cell performance is observed under high strain, demonstrating the mechanical and electrochemical stability of the electrodes.
Acknowledgement: This work was supported by Analog Devices, Inc. Some experiments were performed through use of the core facilities at the Institute for Soldier Nanotechnologies at MIT by the U.S. Army Research Laboratory and the U.S. Army Research Office under contract number W911NF-13-D-0001. This work was also supported by SDBOR competitive grant.

Solar-driven CO₂ conversion needs to reach at least a 10 terawatt scale to have impact on the global fuels market. To realize this unprecedented level of CO₂ fixation using H₂O as electron source, we have adopted a unique method of leveraging atomic layer deposition (ALD) into the seamless constructions for photocatalytic nanotube arrays, which are designed for gas-phase CO₂ reduction using sunlight and water. Specifically, such arrays have a large number of ALD-grown core (Co₃O₄)-shell (SiO₂) nanotubes operating as independent photosynthetic units under ambient conditions. The SiO₂ nanoshell serves as a H⁺ transmitting and O₂ impermeable membrane to spatially separate H₂O oxidation catalysis by Co₃O₄ inside of the tube from CO₂ reduction catalysis by Cu nanoparticles outside of the tube. This inherent product separation geometry allows to address long-standing scientific barriers of optimizing the combinations of photoactive materials (thermodynamic efficiency) and minimizing charge transfer losses and unwanted reactions (quantum efficiency) of artificial photosynthetic systems.
Performance evaluation and stepwise improvement of photocatalytic efficiency of the array include the optimization of the nanowall composition for structural stability, as well as quantitative evaluation and mechanistic understanding of charge-carriers (proton and electron) transfer across the core-shell nanotube wall using atomic resolution imaging, infrared spectroscopy, and electrochemical methods. Photocatalytic efficiency of the fully functionalized nanotube array is measured by time-resolved product monitoring through infrared, optical and mass spectroscopy. This measurement is further underpinned by mechanistic framework derived from parallel photocatalytic and photoelectrochemical studies using cm²-scale planar analogs of the core-shell nanotube walls in H-cell configuration.

Spectroscopic measurements of the gross erosion of crystalline silicon carbide (β-SiC) coatings, fabricated via chemical vapor deposition (CVD), have been measured under D and C ion bombardment using the DIII-D National Fusion Facility [1]. These measurements reveal that the physical sputtering of silicon from these SiC coatings is reduced by a factor of ~10 relative to pure Si material in the low ion impact energy range (50-300 eV) relevant to magnetic fusion devices. This result is consistent with TRIM.SP and MD modeling [2] predicting a suppression of Si physical sputtering from β-SiC due to the enhanced surface binding energy imparted by Si-C chemical bonds. A 2-5x reduction in the C chemical erosion rate from SiC is observed relative to ATJ graphite, the standard wall armor material used in DIII-D and many magnetic fusion experiments. The near-elimination of C chemical sputtering from SiC confirms and extends previous studies on high-flux linear plasma devices [3]. No evidence of Si chemical erosion via silane (Si_xD_y) compounds is observed, but some Si II emission persists below the ion impact energy threshold for Si physical sputtering, suggesting that a chemical release mechanism is present.
SiC samples of 100-200 μm thickness were prepared via CVD on ATJ graphite substrates in the General Atomics Nuclear Technologies & Materials Laboratory. These materials were exposed to energetic (50-300 eV) D ion bombardment up to a fluence of 2x10²³ m^-2. Transient plasma pulses with D ion impact energies of 1-2 keV and heat fluxes up to 10 MW m^-2 were also utilized for thermal shock testing. These materials were tested in the DIII-D tokamak using a removeable sample exposure probe known as the Divertor Materials Evaluation System (DiMES). In-situ measurements of Si and C gross erosion were conducted using line emission spectroscopy and the S/XB method [4]. No macroscopic damage to the SiC coatings was evident upon post-mortem visual inspection and detailed morphology analysis revealed no detectable change to the surface roughness before and after exposure. Microstructural analysis via SEM indicated features with characteristic size < 5 μm were not preserved, potentially due to evaporation and/or viscous flow processes. Energy Dispersive X-Ray Spectroscopy analysis revealed a slight enrichment of the Si surface concentration, relative to C, after plasma exposure. This result is corroborated by the spectroscopically inferred Si gross erosion measurements, which are quantitatively consistent with TRIM.SP physical sputtering predictions under the ad hoc assumption of a 5% Si enrichment fraction of the SiC surface. These results imply potentially favorable erosion properties for crystalline β-SiC under low impact energy hydrogenic ion bombardment charateristic of magnetic fusion energy devices. In conjunction with the robust neutronic properties of SiC at high temperatures, these results motivate serious consideration of SiC as wall armor material in future high power-density magnetic plasma confinement devices.

Work supported by the US Department of Energy under DE-FC02-04ER54698.

[1] https://fusion.gat.com/global/diii-d/home
[2] S. Bringuier et al., Nucl. Mater. Energy 19 (2019) 1-6
[3] J. Westerhout et al., Phys. Scr. T138 (2009) 014017
[4] A Pospieszczyk et al., J. Phys. B: At. Mol. Opt. Phys. 43 (2010) 144017