The recent studies of thermal transport in suspended, supported, and encased graphene just began to uncover the richness of twodimensional phonon physics, which is relevant to the performance and reliability of graphene-based functional materials and devices. Among the outstanding questions are the exact causes of the suppressed basal-plane thermal conductivity measured in graphene in contact with an amorphous material, and the layer thickness needed for supported or embedded multilayer graphene (MLG) to recover the high thermal conductivity of graphite. Here we use sensitive in-plane thermal transport measurements of graphene samples on amorphous silicon dioxide to show that full recovery to the thermal conductivity of the natural graphite source has yet to occur even after the MLG thickness is increased to 34 layers, considerably thicker than previously thought. This seemingly surprising finding is explained by long intrinsic scattering mean free paths of phonons in graphite along both basal-plane and cross-plane directions, as well as partially diffuse scattering of MLG phonons by the MLG-amorphous support interface, which is treated by an interface scattering model developed for highly anisotropic materials. Based on the phonon transmission coefficient calculated from reported experimental thermal interface conductance results, phonons emerging from the interface consist of a large component that is scattered across the interface, making rational choice of the support materials a potential approach to increasing the thermal conductivity of supported MLG.phonon transport | boundary scattering | nanoscale thermal transport | two-dimensional materials | thermal management A s a monoatomic layer of carbon atoms arranged in a hexagonal lattice, single-layer graphene (SLG) is the building block of graphite and carbon nanotubes (CNTs), which can be envisioned as a stack of a large number of graphene layers and rolled-up cylinders of graphene sheets, respectively. Thermal transport in these graphitic materials has intrigued researchers for several decades. The industrial use of graphite in high-temperature or high-heat flux applications motivated a number of initial studies of its thermal properties. These studies have found highly anisotropic thermal transport properties in graphite, where the basal-plane thermal conductivity is among the highest found in solids and nearly two orders of magnitude larger than the value measured along the c-axis (1-3). The recent rediscoveries of CNTs and SLG have expanded the applications of these graphitic nanomaterials for electronic devices, sensors, and light-weight composite materials, among others (4, 5). The performance and reliability of CNT and graphene devices are often closely related to the thermal properties of these nanoscale building blocks, similar to the situation in silicon nanoelectronic devices where localized heating has become a grand challenge (6). Hence, there have been a number of studies of thermal transport in these carbon nanostructures. Some of the studi...

No abstract

On the basis of scanning thermal microscopy (SThM) measurements in contact and lift modes, the low-frequency acoustic phonon temperature in electrically biased, 6.7-9.7 μm long graphene channels is found to be in equilibrium with the anharmonic scattering temperature determined from the Raman 2D peak position. With ∼100 nm scale spatial resolution, the SThM reveals the shifting of local hot spots corresponding to low-carrier concentration regions with the bias and gate voltages in these much shorter samples than those exhibiting similar behaviors in the infrared emission maps.

We report on employing molecular doping to enhance the sensitivity of graphene sensors synthesized via chemical vapor deposition to NH3 molecules at room temperature. We experimentally show that doping an as-fabricated graphene sensor with NO2 gas improves sensitivity of its electrical resistance to adsorption of NH3 molecules by about an order of magnitude. The detection limit of our NO2-doped graphene sensor is found to be ∼200 parts per billion (ppb), compared to ∼1400 ppb before doping. Electrical characterization and Raman spectroscopy measurements on graphene field-effect transistors show that adsorption of NO2 molecules significantly increases hole concentration in graphene, which results in the observed sensitivity enhancement.

Integration of a complementary metal-oxide semiconductor (CMOS) and monolayer graphene is a significant step toward realizing low-cost, low-power, heterogeneous nanoelectronic devices based on two-dimensional materials such as gas sensors capable of enabling future mobile sensor networks for the Internet of Things (IoT). But CMOS and post-CMOS process parameters such as temperature and material limits, and the low-power requirements of untethered sensors in general, pose considerable barriers to heterogeneous integration. We demonstrate the first monolithically integrated CMOS-monolayer graphene gas sensor, with a minimal number of post-CMOS processing steps, to realize a gas sensor platform that combines the superior gas sensitivity of monolayer graphene with the low power consumption and cost advantages of a silicon CMOS platform. Mature 0.18 µm CMOS technology provides the driving circuit for directly integrated graphene chemiresistive junctions in a radio frequency (RF) circuit platform. This work provides important advances in scalable and feasible RF gas sensors specifically, and toward monolithic heterogeneous graphene-CMOS integration generally. 1-3 In these cases, power and size requirements are not critical, and such sensors tend to be large and bulky. But with the rapid expansion and prevalence of newer technologies like smart phones, cloud computing and the Internet of Things (IoT), mobile sensors are recognized as an essential component in future ubiquitous sensor networks. 4,5 The goals of IoT sensor networks require mobile and untethered sensors in quantities that negate any realistic possibility of individual sensor maintenance or battery replacement. The expectation is that the sheer number of future sensor devices, the "things" of IoT, will preclude human maintenance of individual nodes within large sensor networks. The implications for future device production are then twofold: the sensors must operate at low-power, and the cost of each device should be low enough that the expected orders of magnitude increase in sensor nodes is feasible. Much of current gas sensor research is therefore directed at the need for low-cost, low-power portable gas sensors, as well as integration with the technology platform best suited to meet that need: silicon complementary metal-oxide semiconductor (CMOS).Solid-state gas sensors cover a wide range of technologies, from microelectromechanical thermal and mass sensors to optical and chemiresistive sensors. 6,7 Of these, one of the most common is the chemisresistive sensor, whose relatively simple design and operation make it a strong candidate for CMOS integration.