Highest precision limit on possible cosmological variation in

Highest precision limit on possible cosmological variation in the finestructure constant Srđan Kotuš Collaborators: Prof. Michael Murphy Prof. Robert Carswell VARCOSMOFUN’ 16 15. 09. 2016.

Outline Fundamental constants Why? Fine structure constant α How? Recent measurements Stringent limit on variation in α from the z=1. 1508 absorption system towards HE 0515 -4414 quasar 15/09/16 1

Laws of nature Rules that govern our Universe Experiments In our small region of space-time We do not know Why they exist How many laws exists Do we understand them well Today Standard model and General relativity It might not be the whole story 15/09/16 2

Standard model ? http: //www. pha. jhu. edu 15/09/16 3

Scientific process EXPERIMENT PREDICTION MEASUREMENT THEORY 15/09/16 4

Fundamental constants EXPERIMENT PREDICTION MEASUREMENT • FUNDAMENTAL CONSTANTS THEORY 15/09/16 5

Fine-structure constant Introduced by Sommerfeld (1911) How strongly charged particles are bound together HOW PHOTONS INTERACT WITH MATTER! If it was just 4 % different, stars wouldn’t be able to fuse carbon DO NOT WORRY, IT VARIES MUCH LESS THAN THIS! 15/09/16 6

∆α/α measuring methods Different methods Some methods – very precise The quasar absorption method Less precise! Probes large part of the universe Redshift range 0. 2 – 4 No additional assumptions Uzan (2011) 15/09/16 7

Changing strength of EM 15/09/16 8

Quasar light and spectrum http: //astronomy. nmsu. edu/ 15/09/16 9

Quasar constraints Movie Credit: Swinburne Astronomy Productions 15/09/16 10

Quasar constraints Movie Credit: Swinburne Astronomy Productions 15/09/16 11

Quasar constraints Movie Credit: Swinburne Astronomy Productions 15/09/16 12

Quasar constraints Movie Credit: Swinburne Astronomy Productions 15/09/16 13

Quasar constraints Δα/α~ ΔV Movie Credit: Swinburne Astronomy Productions 15/09/16 14

Many Multiplet (MM) method 15/09/16 15

15/09/16 16

15/09/16 17

15/09/16 18

MM method Lower α in Keck sample - Webb at al. (1999) - Murphy et al. (2004) Zero variation in first VLT studies Chand et al. (2008) Significant problems identified by Murphy et al. (2008) Webb et al. (2011) and King et al. (2012) larger value (VLT) When combined with Keck – angular variation Angular variation in α accross the sky 15/09/16 19

Calibration problems Th. Ar wavelength calibration Courtesy of Michael Murphy “Supercalibration” with solar like spectra Distortions not seen in the first attempts (Molaro et al. 2008, Griest et al. 2010 and Whitmore et al. 2010) Rahmani et al. (2013) detected long-range distortions Whitmore & Murphy (2015) found common long-range distortions 15/09/16 20

Calibration problems Whitmore & Murphy (2015) Distortion = Velocity shift changes with wavelength Long-range distortions Short-range distortions Can change in time, still not explained 15/09/16 21

Project aim Most precise reliable measurements of the fine-structure constant variation Brightest quasar in the Southern sky above z = 1 More than 35 h of exposure time with UVES/VLT + HARPS + b. HROS 13 h of new observations High resolving power ~70000 (55000 -93000) Highest signal to noise ~250 pixel-1 echelle quasar spectrum 15/09/16 22

Previous HE 0515 -4414 measurements 15/09/16 23

Spectrum 3 regions Kotuš et al. (2016) - ar. Xiv: 1609. 03860 v 1 15/09/16 24

Main challenges Combine spectra – many exposures Sub-spectra with same slit widths Measure resolving power in each transition in each sub-spectrum Correct velocity shifts Different position of object in the slit Correct the long-range distortions Possibly different between sub-spectra Fitting Data problems 15/09/16 25

Calibration using HARPS m T Velocity shift U /VL S E V ctru e p s HARPS spectrum Wavelength Typical distortions – correspond to ∆α/α ~ 2 ppm Uncertainty in this correction ~ 0. 6 ppm 15/09/16 26

Calibration using ESPRESSO m T Velocity shift U /VL S E V ctru e p s trum ESPRESSO spec Wavelength Correct long-range distortions in existing UVES observations Save telescope time and make more measurements Around 20 UVES spectra convenient for this exists 15/09/16 27

Complexity of fitting ∆α/α bulk property of entire absorption system All components justified by many transitions 15/09/16 28

Fit 15/09/16 29

Measuring ∆α/α Included ∆α/α in the χ2 minimization Kotuš et al. (2016) - ar. Xiv: 1609. 03860 v 1 15/09/16 30

Field status Corrected for long-range distortions Evans et al. (2014) Comparing Solar atlas with adjacent “Solar like” spectrum Kotuš et al. (2016) - ar. Xiv: 1609. 03860 v 1 Immune to long-range distortions Murphy et al. (2016) 15/09/16 31

Our ∆α/α measurement 15/09/16 32

Our ∆α/α measurement 15/09/16 33

Next project UVES Large program – 3 quasar spectra – 13 absorption system 15/09/16 34

Higher resolution spectra Kotuš et al. (2016) ar. Xiv: 1609. 03860 v 1 Comparison with higher resolution spectra No extra structure in HARPS and b. HROS Velocity profile - very narrow closely spaced components No gain from increasing resolution 15/09/16 35

Conclusions New reliable measurement Best current measurement from a single absorption system Unlikely to be matched from 10 -m class telescope spectra Lessons for ESPRESSO Correcting calibration errors in UVES spectra Same object spectrum taken with stable spectrographs Reliable measurements from existing UVES spectra Increasing resolution would not improve measurements significantly Evidence for narrow closely packed components Increasing resolution increases number of pixels and read noise Even lower-resolution mode of ESPRESSO R=120000 a bit too high 15/09/16 36