We are developing millimeter wave (mm-wave) components and circuits based on hydrogen-intercalated graphene. The development covers epitaxial graphene growth, device fabrication, modelling, integrated circuit design and fabrication, and circuit characterizations. The focus of our work is to utilize the distinctive graphene properties and realize new components that can overcome some of the main challenges of existing mm-wave technologies in term of linearity.

We demonstrate high vertical yield InAs1-xSbx (0 < x ≤ 0.18) nanowire arrays grown on InP (111)B substrates by calalyst-free selective-area metal-organic chemical vapor deposition. High antimony composition is achieved by pulsing the arsenic flow to reduce the effective arsenic partial pressure while keeping the antimony partial pressure fixed. This increases the antimony vapor phase composition while allowing the antimony partial pressure to be kept low enough to avoid antimony condensation on the growth mask. InAsSb nanowire arrays show strong emission by photoluminescence at 77 K, covering a wavelength range of 3.77 – 5.08 μm. These results pave the way to engineering optical properties and enabling hybrid integration for nanoscale mid-wavelength infrared optical devices.

The promise of nanocomposites lies in their multi-functionality, the possibility of realizing unique combinations of properties that are not attainable in traditional materials. Ceramic/metal multilayers (CMMs) are one such unique combination that are becoming increasingly popular among researchers today. The idea is to combine the superior properties of ceramics like hardness and strength with favorable properties of metal such as ductility. Materials with these characteristics have potential for engineering applications such as highly efficient gas turbines, aerospace materials, automobiles, protective coatings, etc. Molecular dynamics atomistic simulations were performed to study the scratch behavior of different models of niobium carbide (NbC)-niobium (Nb) multilayers. The layer thicknesses were varied and the coefficient of friction was calculated at various depths of indentation. The deformation mechanisms were investigated to explain the observed mechanical behavior of the models under scratching. Model with the lowest metal/ceramic thickness ratio (2nm NbC/2nm Nb) showed the highest hardness, highest scratch resistance, and also highest friction coefficient. However, this model also showed the highest materials removal rate.

Composite nanostructured foams consisting of a metallic shell deposited on a polymeric core were formed by plating copper via electroless deposition on electrospun polycaprolactone (PCL) fiber mats. The final structure consisted of 1000-nm scale PCL fibers coated with 100s of nm of copper, leading to final core-shell thicknesses on the order of 1000-3000 nm. The resulting open cell, core-shell foams had relative densities between 4 and 15 %. By controlling the composition of the adjuncts in the plating bath, particularly the composition of formaldehyde, the relative thickness of copper coating as the fiber diameter could be controlled. As-spun PCL mats had a nominal compressive modulus on the order of 0.1 MPa; adding a uniform metallic shell increased the modulus up to 2 MPa for sub-10 % relative density foams. A computational materials science analysis using density functional theory was used to explore the effects pre-treatment with Pd may have on the density of nuclei formed during electroless plating.

Electrodeposition in pulse current mode of nickel-molybdenum alloy on a nickel substrate was studied. The range of current density variation from 2 to 9 A/dm2 was investigated. The range of pulse and pause step lengths is from tens to hundreds of milliseconds. SEM-images of applied coatings surfaces are obtained. The method of energy dispersive spectroscopy determined that the molybdenum content in the coating is 21-24 wt%. It was found that under transient pulse mode of electrolysis, with the pulse step corresponding to hundreds of milliseconds, the most rigid and smooth coatings of the electrolytic nickel-molybdenum alloy are obtained from ammonium-citrate electrolyte. It is shown that the percentage of nickel in the alloy does not depend on the electrolysis mode.

Bulk metallic glass and their composites are unique new materials which have superior mechanical and structural properties as compared to existing conventional materials. Owing to this, they are potential candidates for tomorrow’s structural applications. However, they suffer from disadvantages of poor ductility and little or no toughness which render them brittle and they manifest catastrophic failure on the application of force. Their behavior is dubious and requires extensive experimentation to draw conclusive results. In present study, an effort has been made to overcome this pitfall by simulation. A quantitative mathematical model based on KGT theory has been developed to describe nucleation and growth of second phase dendrites from melt in glassy matrix during solidification. It yields information about numerical parameters necessary to understand the behaviour of each individual element in multicomponent sluggish slurry and their effect on final microstructural evolution. Model is programmed and simulated in MATLAB®. Its validation is done by comparison with identical curves reported in literature previously for similar alloys. Results indicate that the effect of incorporating all heat transfer coefficients at macroscopic level and diffusion coefficients at microscopic level play a vital role in refining the model and bringing it closer to actual experimental observations. Two types of hypo and eutectic systems namely Zr65Cu15Al10Ni10 and Zr47.5Cu45.5Al5Co2 respectively were studied. Simulation results were found to be in good agreement with prior simulated and experimental values.

Polymer imprint thermal mapping (PITM) is a high-resolution thermal mapping technique that is especially valuable for nanoscale plasmonic devices. PITM leverages a ∼50 nm polymer film coating that crosslinks irreversibly with temperature, which records the peak temperature rise of the surface in the local, linear reduction of polymer film thickness. Using AFM to measure topography before and after heating, but not during operation, PITM sidesteps plasmonic artifacts seen in other near-field thermometries, where the probe tip disturbs and is heated directly by the near- and far-field radiation around the plasmonic device. This is notably troublesome for characterizing heat-assisted magnetic recording (HAMR) heads for next-generation hard disk drives. HAMR heads use near-field transducers (NFTs) to focus light on a magnetic media, heating a nanoscale region to its Curie temperature to enable magnetic writing. The PITM proof-of-concept was introduced at The Magnetic Recording Conference (TMRC) in 2015: here, we present a mature technique capable of benchmarking finite-element thermal simulations of nanoscale devices.

Thermal characterization of nano-featured devices is a critical challenge for the development of high performance devices. Although far-field thermoreflectance imaging is limited in spatial resolution by the optical diffraction limit, it is more amenable to absolute temperature calibration of plasmonic devices than existing near-field scanning probe tip methods. We have built an advanced thermoreflectance microscope capable of 50 ns time-resolved, diffraction-limited temperature imaging that can account and correct for thermal expansion, sample drift, numerical aperture, and polarization induced variations in the apparent thermoreflectance coefficient of nanoscale structures. We developed a per-pixel transient calibration technique using this microscope to measure the absolute temperature of an operating heat-assisted magnetic recording (HAMR) head, including features as narrow as 200 nm. The resulting temperature information can be used to experimentally validate numerical models in the design process of such plasmonic devices.

Thermal management is a growing challenge for electronics packaging because of increased heat fluxes and device miniaturization. Thermal interface materials (TIMs) are used in electronic devices to transfer heat between two adjacent surfaces. TIMs need to exhibit high thermal conductivity and must be soft to minimize thermal contact resistance. Polymers, despite their relative softness, suffer from low thermal conductivity (∼0.2 W/m-K). To overcome this challenge, we infiltrate nanoporous anodic aluminum oxide (AAO) templates with molten polymer to fabricate large area arrays of vertically aligned polymer nanofibers. Nanoscale confinement effects and flow induced chain elongation improve polymer chain alignment (measured using polarized Raman spectroscopy) and thermal conductivity (measured using the photoacoustic method) along the fiber’s long axis. Conjugated poly(3-hexylthiophene-2,5-diyl) (P3HT) and non- conjugated polyethylene (PE) of various molecular weights are explored to establish a relationship between polymer structure, nanofiber diameter, and the resulting thermal conductivity. In general, thermal conductivity improves with decreasing fiber diameter and increasing polymer molecular weight. Thermal conductivity of approximately 7 W/m-K was achieved for both the ∼200 nm diameter HDPE fibers and the 100 nm diameter P3HT fibers. These results pave the way for optimization of the processing conditions to produce high thermal conductivity fiber arrays using different polymers, which could potentially be used in thermal interface applications.

For heat-assisted magnetic recording (HAMR) heads, a major reliability limiter is the peak near-field transducer (NFT) temperature. Since the NFT is nanoscale, heat sinking is controlled by materials and interfaces within a few 100 nm of the NFT. Heat sinks can be metallic to take advantage of the 10x-100x higher thermal boundary conductance (TBC) of metal/metal interfaces, versus nonmetal interfaces. Oxide formation at these interfaces can greatly decrease the TBC and contribute to NFT failure. Likewise, the thermal resistance of material between the NFT and media recording layer greatly influences the NFT operating temperature. Here we use pump-probe thermoreflectance techniques (FDTR, TDTR) to study metal-metal interfaces and detect partial oxidation of a buried metallic thin film, as well as evaluate the interface thermal conductance of amorphous-amorphous interfaces in a film stack representative of a HAMR head-media interface.

Thermoelectric materials have been of interest for several decades due to their ability to recapture waste heat of various systems and convert it to useful electricity. One method used to improve the thermoelectric efficiency of a material is to reduce the lattice thermal conductivity (k p ) while not affecting the other properties. In order to reduce the k p of the material, this paper introduces silicon (Si) nanoparticles (NPs) in Mg2Si to manipulate phonon scattering and mean free path. A series of simulations is performed with the metal silicide thermoelectric material MgxSix. The objective of this work is two-fold: 1) to determine the optimal Si nanoparticle (NP) concentration and 2) to determine the optimal MgxSix stoichiometry for minimizing the k p of the system. It should be noted, however, that the assumed reduction in thermal conductivity is only a result of reduced phonon transport and that minimal impact is made on the transport of electrons. Interestingly, the uniform off-stoichiometry (49.55 atomic percent (a/o) Si) sample of MgxSix resulted in a reduction of k p of 84.62 %, while the Si NP sample, with matching a/o Si, resulted in a reduction of k p of 78.82 %.

Pristine graphene has low thermoelectric performance due to its ultra-high thermal conductivity and a low Seebeck coefficient, the latter of which results from the zero-band gap of graphene. To improve the thermoelectric performance of graphene-based materials, various methods have been proposed to open a band gap in graphene. Graphene antidot lattices is one of the most effective methods to reach this goal by patterning periodic nano- or sub-1-nm pores (antidots) across graphene. In high-porosity graphene antidot lattices, charge carriers mainly flow through the narrow necks between pores, forming a comparable case as graphene nanoribbons. This will open a geometry-dependent band gap and dramatically increase the Seebeck coefficient. The antidots also strongly scatter phonons, leading to a dramatically reduced lattice thermal conductivity to further enhance the thermoelectric performance. In computations, the thermoelectric figure of merit of a graphene antidot lattices was predicted to be around 1.0 at 300 K but experimental validation is still required. The electrical conductivity and Seebeck coefficient of graphene antidot lattices on various substrates including SiO2, SiC and hexagonal boron nitride were measured. The antidots were drilled with a focused ion beam or reactive ion etching.

The thermal properties of organic membranes attract much attention due to the fact that heat dissipation in electronic devices limits their functionality and reliability. Here, we enhance the thermal conductivity of polyvinyl alcohol (PVA) membrane using nano-fibers fabricated by electrospinning. Measured by the 3-Omega method, the results show that the effective thermal conductivity of the electrospinning membranes (with/without Cu nanoparticles) are as high as 0.7 W/m-K at room temperature which is as twice as the value of thermal conductivity of amorphous spin-coated PVA membrane (0.35 W/m-K). The mechanism of enhancement are that, compared with amorphous membrane, the phonon scattering is attenuated and the crystallinity is improved in the electrospinning process. Our studies bring new insights in designing new kind of membrane with high thermal conductivity.

Due to the low degree of contact area and weak interfacial adhesion between CNTs and the growth substrate (Cu), large thermal contact resistance is the largest challenge preventing the use of vertically aligned CNTs (VACNTs) as a thermal interface material (TIM). Although significant research has been done regarding the growth of CNTs on reactive substrates by using an appropriate buffer layer in this group’s previous work, there are many unanswered questions associated with using VACNTs as a thermal interface material beyond synthesis. This effort extends prior work on carbon nanotube growth, by concentrating on ways to evaluate/measure CNT-based nanocomposite thermal resistance. In this study, with the use of a laser flash measurement system, the influence of CNT array properties (layer height and density) on the thermal diffusivity and thermal resistance of the CNT composite were investigated. Test results identify a correlation between the CNT array density/thickness and its thermal resistance.

We report an experimental study on the terahertz frequency dielectric properties of manganese cobalt ferrites (MnxCo1−xFe2O4) and nickel cobalt ferrites (NixCo1-xFe2O4) with three different stoichiometry each, x=0.3, x=0.5 and 0.7. Particularly, we present a comparison and discussion of the terahertz frequency indices of refraction of these two ferrites compositions. MnxCo1−xFe2O4 and NixCo1-xFe2O4 pellets with different Mn/Co and Ni/Co ratios (x=0.3, x=0.5 and x=0.7) were prepared by state-of-the-art ceramic processing. The morphology and chemical homogeneity of these ferrites were characterized by energy dispersive x-ray spectroscopy. We observed that the indexes of refraction for manganese cobalt ferrites are 3.22, 3.71 and 3.67 for ratios of 0.3, 0.5 and 0.7, respectively. In the case of nickel cobalt ferrite, the indexes of refraction are 3.53, 3.57 and 3.47 for ratios of 0.3, 0.5 and 0.7 respectively. We notice a substantial difference in the index of refraction for the Mn0.3Co0.7Fe2O4. This difference may be correlated to a secondary phase formed in this sample.

Zinc oxide/silicon nanowires (ZnO/SiNWs) nanocomposites is a promising material for heterojunction solar cells. They combine the low-reflectivity of SiNWs, where photogenerated charge carriers are produced and harvested, and the high transparency of ZnO, which serves as a functional transparent conductive electrode. In this paper, we present a study of the anti-reflective properties of ZnO/SiNWs core-shell nanostructures. SiNWs were fabricated by a two-step metal-assisted chemical etching and coated with ZnO by electrochemical deposition. Particularly, the change in the specular reflectance of ZnO/SiNWs nanocomposites as a function of thermal annealing temperature under ambient atmosphere is investigated. First, it was shown that the reflectance in the wavelength range of 400-1000 nm of as-synthesized ZnO/SiNWs nanocomposites increases when compared to the bare SiNWs formed from Si wafers with resistivity of 0.3 and 12 Ω∙cm by an 0.51 % and 0.47 %, respectively. Second, it was found that annealed ZnO/SiNWs had a 0.26 % and 0.17 % lower reflectance in the wavelength range of 400-1000 nm than as-synthesized ZnO/SiNWs and yet higher than bare SiNWs. Potential causes such results are discussed in the context of existing literature.