Infrared ( 20-120 and 900-1100 cm(-1)) Faraday rotation and circular dichroism are measured in high- T(c) superconductors using sensitive polarization modulation techniques. Optimally doped YBa2Cu3O7 thin films are studied at temperatures in the range ( 15

We measure the temperature and frequency dependence of the complex Hall angle for normal state YBa2Cu3O7 films from dc to far-infrared frequencies (20-250 cm −1 ) using a new modulated polarization technique. We determine that the functional dependence of the Hall angle on scattering does not fit the expected Lorentzian response. We find spectral evidence supporting models of the Hall effect where the scattering ΓH is linear in T, suggesting that a single relaxation rate, linear in temperature, governs transport in the cuprates.The normal state Hall effect in cuprate superconductors exhibits an anomalous temperature dependence that cannot be explained using conventional transport theory for metals. According to the simple Drude model, the resistivity of a metal and the cotangent of its Hall angle cot(θ H ) = σxx σxy , should share the same temperature dependence, both proportional to the scattering rate of the charge carriers. However, the normal state resistance of cuprate superconductors is linear with temperature, ρ ∼ T , while the Hall angle has a robust cot(θ H ) ∼ T . This apparent duality of scattering rates characterizes the anomalous Hall transport in the cuprates. Several theories approached the problem assuming that two scattering rates were in fact involved, beginning with the spin-charge separation model of Anderson wherein the two species of quasiparticles each relaxed at the different observed rates.[5] Subsequent explanations focused either on alternative non-Fermi liquid mechanisms [6,7] or on the effects of k-space scattering anisotropies [8][9][10][11]. The common feature of all the above theories is a dominant term that is linear in the scattering rate, cot(θ H ) ∼ γ H . In contrast, a recent theory by Varma and Abrahams [12] treats anisotropic scattering in a marginal Fermi-liquid and predicts a square-scattering response, cot(θ H ) ∼ γ 2 H . These different models can be distinguished at finite frequency. The linear-and square-scattering models correspond to Lorentzian and square-Lorentzian spectral responses respectively, and although Hall experiments have been performed at finite frequencies [13][14][15], this paper is the first to study both temperature and frequency dependence of the Hall response in a frequency range that discerns a lineshape and extrapolates to the dc limit.We begin by reviewing the concept of a frequency dependent Hall angle [16] using the Drude model as an example of a Lorentzian response. All parameters are implicitly spectral, i.e. θ H = θ H (ω), and in the present case of strong scattering, tan(θ) ≃ θ << 1. Quasiparticles circling at the cyclotron frequency ω *

No abstract

No abstract

Microfluidic chips require connections to larger macroscopic components, such as light sources, light detectors, and reagent reservoirs. In this article, we present novel methods for integrating capillaries, optical fibers, and wires with the channels of microfluidic chips. The method consists of forming planar interconnect channels in microfluidic chips and inserting capillaries, optical fibers, or wires into these channels. UV light is manually directed onto the ends of the interconnects using a microscope. UV-curable glue is then allowed to wick to the end of the capillaries, fibers, or wires, where it is cured to form rigid, liquid-tight connections. In a variant of this technique, used with light-guiding capillaries and optical fibers, the UV light is directed into the capillaries or fibers, and the UV-glue is cured by the cone of light emerging from the end of each capillary or fiber. This technique is fully self-aligned, greatly improves both the quality and the manufacturability of the interconnects, and has the potential to enable the fabrication of interconnects in a fully automated fashion. Using these methods, including a semi-automated implementation of the second technique, over 10,000 interconnects have been formed in almost 2000 microfluidic chips made of a variety of rigid materials. The resulting interconnects withstand pressures up to at least 800psi, have unswept volumes estimated to be less than 10 femtoliters, and have dead volumes defined only by the length of the capillary.