BICEP Array
BICEP Array will continue the BICEP/Keck program of CMB polarization observations, extending its search for inflationary gravitational waves (IGW). BICEP Array will measure the polarized sky in five frequency bands to reach an ultimate sensitivity to the amplitude of IGW of σ(r) ~< 0.003, extrapolating from achieved performance, and after conservatively accounting for the Galactic dust, Galactic synchrotron and CMB lensing foregrounds. This measurement will be a definitive test of slow-roll models of inflation, which generally predict a gravitational-wave signal above approximately 0.01.
Between the 2017 and 2018 observing seasons we will start to replace Keck Array with BICEP Array. Since the BICEP3 class receivers are physically larger than the Keck Array receivers a new telescope mount is required. The first BICEP Array receiver (installed for the 2018 season) will therefore be a new design operating in a band centered at 35 GHz. Before the 2019 observing season we will install BICEP Array receivers at 95 GHz and 150 GHz. Perhaps surprisingly our projection calculations indicate that by this point it will be optimal to make further measurements at 150 GHz even though the dust signal is much stronger there than at 95 GHz. Finally, before the 2020 season we will install a 220/270 GHz dual band receiver, where the detectors at the two frequencies are interleaved in a "checkerboard" pattern. This will be by far the most technically challenging of the new receivers due to the very high channel count (≈ 19, 000 detectors) which necessitates consideration of new readout technology.
BICEP Array will continue the use of Caltech/JPL's successful dual-polarization, antennacoupled detector array technology as described in our recent Detectors Paper [45]. Hundreds of camera pixels are photolithographed directly on 4" or 6" Si wafers, combining beam-defining slot-antenna networks, frequency-band-defining filters, and transition-edge sensors (TES). These detectors have led to published scientific results from fielded 150 GHz and 95 GHz detectors, and have been successfully acquiring science data in the field at 220 GHz since early 2015. The 270 GHz band detectors have been lab tested, and 35 GHz prototypes have been fabricated and await lab testing. The 35 GHz detectors differ in that detector noise is more significant with respect to photon noise, but the G's are within demonstrated parameters, and the NEP's are the same as detectors flown in the SPIDER experiment. The Keck Array focal planes use 4×4" tiles, BICEP3 uses 20 × 4", and BICEP Array will use 12 × 6" as illustrated in these figures below.
