Highly accurate transition frequencies in the H 2 Lyman and Werner absorption bands

The Journal of Chemical Physics

2014

114

Precision measurement of the rotational energy-level structure of the three-electron molecule He 2 + Transitions in atoms and molecules provide an ideal test ground for constraining or detecting a possible variation of the fundamental constants of nature. In this perspective, we review molecular species that are of specific interest in the search for a drifting proton-to-electron mass ratio μ.In particular, we outline the procedures that are used to calculate the sensitivity coefficients for transitions in these molecules and discuss current searches. These methods have led to a rate of change in μ bounded to 6 × 10 −14 /yr from a laboratory experiment performed in the present epoch. On a cosmological time scale, the variation is limited to | μ/μ| < 10 −5 for look-back times of 10-12 × 10 9 years and to | μ/μ| < 10 −7 for look-back times of 7× 10 9 years. The last result, obtained from high-redshift observation of methanol, translates intoμ/μ = (1.4 ± 1.4) × 10 −17 /yr if a linear rate of change is assumed.

Section: A Transitions In Molecular Hydrogen and Carbon Monoxidementioning

confidence: 93%

Perspective: Tipping the scales: Search for drifting constants from molecular spectra

Jansen

Bethlem

The Journal of Chemical Physics

2014

114

“…The setup was used for an extensive experimental investigation on the spectroscopy of the Lyman and Werner bands of H 2 calibrating in total 138 rotationally resolved lines at an accuracy varying from 0.004 to 0.011 cm À1 , depending on wavelength range [88]. The experimental setup employed for the H 2 laboratory spectroscopic studies is presented in Fig.…”

Section: H 2 Spectroscopy In the Laboratorymentioning

confidence: 99%

On a possible variation of the proton-to-electron mass ratio: H2 spectra in the line of sight of high-redshift quasars and in the laboratory

Journal of Molecular Spectroscopy

Buning

Eikema

et al. 2007

124

Recently the finding of an indication for a decrease of the proton-to-electron mass ratio l = m p /m e by 0.002% in the past 12 billion years was reported in the form of a Letter [E. Reinhold, R. Buning, U. Hollenstein, P. Petitjean, A. Ivanchik, W. Ubachs, Phys. Rev. Lett. 96 (2006) 151101]. Here we will further detail the methods that led to that result and put it in perspective. Laser spectroscopy on molecular hydrogen, using a narrow-band and tunable extreme ultraviolet laser system at the Laser Centre Vrije Universiteit Amsterdam, results in transition wavelengths of spectral lines in theWerner band systems at an accuracy of (4-11) · 10 À8, depending on the wavelength region. This corresponds to an absolute accuracy of 0.000004-0.000010 nm. A database of 233 accurately calibrated H 2 lines is presented here for future reference and comparison with astronomical observations. Recent observations of the same spectroscopic features in cold hydrogen clouds at redshifts z = 2.5947325 and z = 3.0248970 in the line of sight of two quasar light sources (Q 0405À443 and Q 0347À383) resulted in 76 reliably determined transition wavelengths of H 2 lines at accuracies in the range 2 · 10 À7 to 1 · 10 À6 . Those observations were performed with the Ultraviolet and Visible Echelle Spectrograph at the Very Large Telescope of the European Southern Observatory at Paranal, Chile. A third ingredient in the analysis is the calculation of an improved set of sensitivity coefficients K i , a parameter associated with each spectral line, representing the dependence of the transition wavelength on a possible variation of the proton-to-electron mass ratio l. The new model for calculation of the K i sensitivity coefficients is based on a Dunham representation of ground state and excited state level energies, derived from the most accurate data available in literature for the X 1 R þ g ground electronic state and the presently determined level energies in the B 1 R þ u and C 1 P u states. Moreover, the model includes adiabatic corrections to electronic energies as well as local perturbation effects between B and C levels. The full analysis and a tabulation of the resulting K i coefficients is given in this paper. A statistical analysis of the data yields an indication for a variation of the proton-to-electron mass ratio of Dl/l = (2.45 ± 0.59) · 10 À5 for a weighted fit and Dl/l = (1.99 ± 0.58) · 10 À5 for an unweighted fit. This result, indicating the decrease of l, has a statistical significance of 3.5r. Mass-variations as discussed relate to inertial or kinematic masses, rather than gravitational masses. Separate treatment of the data gives a similar positive result for each of the quasars Q 0405À443 and Q 0347À383. The statistical analysis is further documented and possible systematic shifts underlying the data, with the possibility of mimicking a non-zero Dl/l value, are discussed. The observed decrease in l corresponds to a rate of change of d lnl/dt = À2 · 10 À15 per year, if a linear variation with time is assumed. Expe...

“…From an analysis of combination differences between R J and P J 2 lines, it follows that the data are more consistent than estimated; hence, the stated uncertainty represents a conservative estimate. Further details on the experimental methods and a critical evaluation of the uncertainties are given by Philip et al [17].…”

mentioning

confidence: 99%

Highly AccurateH2Lyman and Werner Band Laboratory Measurements and an Improved Constraint on a Cosmological Variation of the Proton-to-Electron Mass Ratio

Reinhold

2004

Phys. Rev. Lett.

Transition wavelengths on a large set of H 2 Lyman and Werner band spectral lines have been obtained at an accuracy of 5 10 ÿ8 , using a narrow band tunable extreme ultraviolet laser. The data are used to determine a constraint on a possible cosmological variation of the proton-to-electron mass ratio ( M p =m e ) from a comparison with highly redshifted spectral data of quasistellar objects, yielding a fractional change in the mass ratio of = ÿ0:5 3:6 10 ÿ5 (2 ), which would correspond to a temporal change of d=dt = ÿ0:4 3:0 10 ÿ15 per year (2 ) if a linear cosmological expansion model is assumed. DOI: 10.1103/PhysRevLett.92.101302 PACS numbers: 98.80.Es, 14.20.Dh, 33.20.Ni, 98.62.Ra The search for a cosmological variability of fundamental physical constants has become an active field of research now that accurate spectroscopic data can be obtained from quasars that have emitted their radiation more than ten billion years ago. A comparison between laboratory data, obtained at the highest accuracy in the modern epoch, and analysis of corresponding redshifted spectra from distant objects allows for stringent constraints on variability of the physical constants underlying the spectra. One such issue is that of the fine structure constant , which was claimed to vary based on observations of quasar spectra [1]. Another is that of the proton-to-electron mass ratio M p =m e , a dimensionless parameter equalling 1836.152 672 61(85) [2]. The idea that fundamental physical constants might not be constant on a cosmological time scale was pioneered by Dirac, Teller, Gamov, and others; the observational status and theoretical perspectives were recently discussed by Uzan [3], citing all relevant publications.The spectrum of molecular hydrogen with the prominent Lyman and Werner band systems may be used to detect a change in over time, because each individual spectral line depends in a different way on the mass ratio, as was pointed out by Thompson [4]. In recent years, several observations have been made on the Lyman and Werner bands in quasars, taking advantage of the fact that Earth's atmosphere is transparent for these spectral regions at high z, and that ground-based echelle-grating spectrometers can be employed in high-resolution studies [5][6][7][8][9][10]. In comparisons, the laboratory data of the Lyman and Werner band tables of the Meudon group were used [11,12], which represent an extensive and accurate database. Nevertheless, it was emphasized that improved spectroscopic data are desired [8], because the statistical errors from astronomical and laboratory data contributed equally in the estimates of = , the fractional variation of . This is illustrated by the fact that Ivanchik et al. [8] find rather different constraints on = , when comparing quasar data with different laboratory data sets [11][12][13]. Here we report on laser-based laboratory determinations of Lyman and Werner band spectral lines at an accuracy improved by more than an order of magnitude; the resulting laboratory wavelengths 0 i can be consid...