Polarization Angle Calibration And B-Mode Characterization With The Bicep And Keck Array CMB Telescopes

Abstract

Since its discovery in 1964, the Cosmic Microwave Background (CMB) has led to widespread acceptance of the Big Bang cosmological paradigm as an explanation for the evolution of the Universe. However, this paradigm does not explain the origin of the initial conditions, leading to such issues as the “horizon problem” and “flatness problem.” In the early 1980’s, the inflationary paradigm was introduced as a possible source for the initial conditions. This theory postulates that the Universe underwent a period of exponential expansion within a tiny fraction of a second after the beginning. Such an expansion is predicted to inject a stochastic background of gravitational waves that could imprint a detectable B-mode (curl-like) signal in the polarization of the CMB. It is this signal that the family of telescopes used by the BICEP1, BICEP2, and Keck Array collaborations were designed to detect. These telescopes are small aperture, on-axis, refracting telescopes. We have used the data from these telescopes, particularly BICEP2 and the Keck Array, to place the tightest constraints, as of March 2016, on the tensor-to-scalar ratio of the CMB of r_{0.05} < 0.07. In this dissertation, we provide an overview of the Keck Array telescopes and analysis of the data. We also investigate, as the main focus of this dissertation, a device we call the Dielectric Sheet Calibrator (DSC) that is used to measure the polarization angles of our detectors as projected on the sky. With these measurements, we gain the potential to separate the polarization rotation effects of parity-violating physics, such as cosmic birefringence, from a systematic uncertainty on our detectors’ polarization angles. Current calibration techniques for polarization sensitive CMB detectors claim an accuracy of ±0.5°, which sets a limit for determining the usefulness of the DSC. Through a series of consistency tests on a single Keck Array receiver, we demonstrate a statistical uncertainty on the DSC measurements of ±0.03° and estimate a systematic uncertainty of ±0.2°. which meets the minimum goal. We also conclude that there is no conflict between the DSC-derived polarization angles of this single receiver and the rotation derived from that receiver’s CMB data under the hypothesis of no cosmic birefringence.