Arxiv:1912.01335V4 [Physics.Atom-Ph] 14 Feb 2020

Home , Hartree

arXiv:1912.01335v4 [physics.atom-ph] 14 Feb 2020 esvl icse nteltrtr,se o xml,the example, for [ see, review literature, the in discussed tensively cl eeatt esrmn uha nteatomic the time in a as on such measurement variations [ a fast experiments the to having in relevant constants vary scale to the only allowed also of whether We be case of may question constants modified. contentious dimensionless and slow often of revisited the case the be the discuss of to for part established need were a of variations that some of rules because time basic important the is response This the vari- system. the than experimental of faster of time-scale variation is the apparent a where ation with situation provide deal a We in to constants how arise. of may recipe “constants” general of appar- variations to which ent in is context article, [ beyond-the-standard-model others present by the the discussions given earlier of a the during goal following constant clarify, The fact, in run. could be, experimental “constants” to the assumed safely so be measurement, assumed that individual than usually an longer much was of scales it time at literature, occur variations previous that the in ever, Introduction. 1 ,adrfrne hri swl s[ as well as therein references and ], eateto hsc,Uiest fClfri tBerkele at California of University Physics, of Department eesugNcerPyisIsiueo R KrhtvIns “Kurchatov NRC of Institute Physics Nuclear Petersburg es,oemysyta h xeietlsga eed nth on depends signal experimental the that ( say may one sense, iesols uniyi,freape h ratio the example, experimental for o or is, number species studied constants quantity the the the dimensionless of of However, response variations the fast of studied. for scale allow is fund we other when levels increases to atomic sensitive of less structure usually is spectroscopy Atomic fvrain ftefiesrcueconstant structure experimen These fine the constants. of fundamental variations of of variation the on m 12 t eesugEetoehia nvriy“EI,Prof. “LETI”, University Electrotechnical Petersburg St. rcso pcrsoyo tm n oeue losoeto one allows molecules and atoms of spectroscopy Precision e , aitoso cntns aebe ex- been have “constants” of Variations emot nttt an,Jhne uebr University Gutenberg Johannes Mainz, Institute Helmholtz nti example). this in 13 emot nttt an,Jhne uebr University Gutenberg Johannes Mainz, Institute Helmholtz emot nttt an,Jhne uebr University Gutenberg Johannes Mainz, Institute Helmholtz IA ainlIsiueo tnad n ehooy n D and Technology, and Standards of Institute National JILA, ,o h xeietwt resonant- with experiment the or ], colo hsc,Uiest fNwSuhWls yny205 Sydney Wales, South New of University Physics, of School ataprn siltoso udmna constants fundamental of oscillations apparent Fast ezanIsiueo cec,Rhvt sal7610001 Israel Rehovot, Science, of Institute Weizmann nvriyo ooao ole,Clrd 00,USA 80309, Colorado Boulder, Colorado, of University eateto atcePyisadAstrophysics, and Physics Particle of Department 2 – itrV Flambaum V. Victor 9 .How- ]. ihi .Kozlov G. Mikhail inso Antypas Dionysios 10 Dtd a 2019) May (Dated: mtyBudker Dmitry , ia Perez Gilad 11 α u Ye Jun ], n h lcrnt rtnms ratio mass proton to electron the and m oiainfr n infiatyepnso h analysis [ the on in expands presented significantly and for, motivation edi siltn ttefeunycoeto close frequency the at oscillating is field asatna [ antennae mass nryi ml oprdt h etenergy rest kinetic field the that to DM means compared This galactic small is non-relativistic. The energy be to space. known as the is exist filling but fields condensate, vacuum classical form necessarily not do any and masses nature. quark of the constants of elec- changes other the also or constant but structure mass, fine tron the to which confined nature not of are of constants time-varying description other enables for formalism searches relevant presented the The of description physics. the for Lagrangian relativistic akMte D)i soitdwt lrlgtscalar [ ultralight Refs. with example, for associated (see, models, is Higgs fields some (DM) scalar In Matter the condensate. Dark with vacuum forms interaction which the field, by generated are edwt emoslast emi hi Lagrangian, their in term a to leads fermions with field where flgt and light, of e / h asso h atce ntesadr oe (SM) model standard the in particles the of masses The h m e paau sd nti ae h relevant the case, this In used. apparatus i ,Bree,Clfri 42-30 USA 94720-7300, California Berkeley, y, hr fs”i eemndb h time the by determined is “fast” where , m mna osat,uls h hyperfine the unless constants, amental and oo t.5 936S.Petersburg St. 197376 5, Str. Popov iue,Gthn 830 usaand Russia 188300, Gatchina titute”, stems ftesaa particle, scalar the of mass the is saetpclyitrrtdi terms in interpreted typically are ts aito fdmninu constants dimensionful of variation e erhfradt u tign limits stringent put to and for search 59 an,Grayand Germany Mainz, 55099 , osbedmninesconstants dimensionless possible f h m h 52 an,Germany Mainz, 55128 , 13 59 an,Germany Mainz, 55099 , e sPakscntn.Itrcino uha such of Interaction constant. Plank’s is i .Tenwaayi sdn sn fully a using done is analysis new The ]. 14 prmn fPhysics, of epartment stetm vrg.I this In average. time the is .Tepeetwr rvdsafull a provides work present The ]. ,Asrlaand Australia 2, 1 µ – 7 = , 9 m , e 11 /m ) hs fields These ]). p c . mc ν stespeed the is = 2 n the and mc 2 /h , 2 which looks like an oscillating mass term. Because of these vibrational and rotational energies, In this scenario of oscillating DM field linearly coupled molecular spectra are sensitive to the mass ratio µ. Rel- to fermions, the particles acquire apparent modifications ativistic corrections again introduce an α dependence: 2 to their masses [2, 5–7, 9, 11], which oscillates at the fre- Cv = Cv,0 + Cv,1(αZ) + ... and similarly for Cr. This quency ν. The amplitude of these oscillations depends on dependence comes about because the molecular potential the local density of the scalar field. If, for example, the and the inter-nuclear distance (that enters the moment apparent mass of the electron me is modified in such a of inertia and thus the rotational energy) depend on the way, this must affect the spectra of atoms and molecules. electronic wavefunctions and thus on αZ. As long as this additional mass-like term appears from For completeness, we need to mention that the hyper- the interaction with the cosmological field, it does not vi- fine structure of atomic and molecular levels is sensitive olate conservation of energy, though the energy of atoms to the nuclear magnetic and quadrupole moments, which is changing.1 If the scalar field is also coupled to the elec- depend on other fundamental constants. With this ex- tromagnetic field, this generally leads to the variation in ception, all the ratios of atomic and molecular transition the strength of the electromagnetic coupling character- frequencies are sensitive only to the values of two funda- 2 e mental constants, namely, α and µ.3 ized by the fine structure constant α = ~c [2–6, 9, 11]. As a result, the fine structure constant also acquires os- Experimental consequences. Let us first assume slow cillating components. For an example of a model with variation of the “constants” on all time scales relative to oscillating me and α, see e.g. [5, 7, 9]. Below we discuss a measurement. Many spectroscopic experiments use op- how such effects can be observed in precision spectro- tical resonators (cavities). The latest state-of-the-art op- scopic experiments. tical resonators use crystalline material, instead of amor- In the non-relativistic approximation, the energy of phous low-expansion glasses, for cavity spacers [17, 18]. any electronic level in an atom is proportional to the The length L of such a cavity depends on the lattice con- atomic unit of energy, Hartree: stant of the material its spacer is made of. The latter, in turn, is proportional to the Bohr radius m e4 E = e 27eV , (1) ~2 H ~2 ≈ r0 = 2 . (4) mee where e is elementary charge, and we write the analyti- cal expression in Gaussian units. In this approximation, The resonant frequency of such cavity is proportional to all atomic transition frequencies are also proportional to c/r0: E and their ratios do not depend on fundamental con- H c m e2c E stants [15, 16]. When relativistic corrections are taken ν = C = C e = C H , (5) cav c c ~2 c ~ into account, the energies acquire a dependence on the r0 α fine structure constant: 2 where Cc = Cc,0 + Cc,1(αZ) + ... We see that the ra- E = E C + C (αZ)2 + ... , (2) tio of atomic transition frequency νat to νcav to a first at H 0 1 approximation is proportional to α: where Z is the number of protons in the nucleus. For δ (νat/νcav) δα neutral atoms, the coefficients Ci are of the order of unity = [1 + (αZ)] . (6) νat and depend on the quantum numbers of the level. For ( /νcav ) α O light atoms, αZ 1, and the dependence of the energies ≪ If the constants are rapidly oscillating, the spectra we on α is weak; however, it becomes significant for heavy study will depend on some average values of the constants elements with Z 100. ≈ and the corresponding averaging time depends on the Electronic energy of light molecules is also proportional response time of the atoms/molecules and the apparatus to EH , but now there are also vibrational and rotational we use. For an atom, the response time depends on the energies Evib and Erot, which depend on the electron to 2 lifetime of the level τat and the width of the transition proton mass ratio µ = me/mp:

1/2 Evib = CvEH µ , Erot = CrEH µ . (3) 3 Strictly speaking, the ratios of atomic frequencies depend on all fundamental constants. However, their sensitivity to other fundamental constants is orders of magnitude smaller. For example, the finite nuclear size leads to the “volume shifts” of atomic 1 −5 In general, all varying fundamental “constants” must be properly levels, typically on the scale 10 EH . The size of the nucleus described as dynamic scalar fields. Indeed, a variation of “con- depends on the strong coupling constant. Thus, the sensitivity stants” leads to a change of energy of a system (e.g., an atom), of atomic energy levels to the variations of the strong coupling which, assuming energy conservation, must be compensated by constant is suppressed by roughly five orders of magnitude. The the energy taken up by the field. advent of laser spectroscopy of a low-energy nuclear transition in 2 229 More generally, the ratio of me to the nucleon mass or the strong Th is expected to be a game-changer with greatly enhanced interaction scale ΛQCD may be considered. sensitivity to nuclear parameters [19, 20]. 3

Γ. For a resonator with a finesse the response time is Fµν stands for the corresponding field strength), a lep- F T τcav 1 L/c. ton doublet, L = (ν ,e ), with ν electron-neutrino , ∝F e e L e For a resonator there is also another relevant time. and eL,R left-handed and right-handed electron fields, T This is the time τcav,2 during which the length L may the Higgs field written in unitary gauge as H = (0,h + adjust to the changing value of the atomic length scale v)/√2 , with h being the celebrated Higgs boson, and r0. We can estimate τcav,2 in terms of the speed of sound v 246GeV being the Higgs vacuum expectation value ≃ in the material v [21], τcav 2 L/v . If the finesse is (VEV). In addition we have a new scalar field ϕ, the sin- s , ≈ s < c/vs, then τcav = τcav,2 > τcav,1. A more accurate glet of the SM gauge interactions. The relevant part of analysisF has to account for other vibrational modes of the Lagrangian is (for more detail see, for Ex. [26, 27]): the cavity [14], but for the estimates one can still use τcav L/vs. 1 µν 1 µ 2 2 free = Fµν F ∂µϕ ∂ ϕ m ϕ As≈ an example, consider the experiment [13] where the L −4α − 2 − frequency of the 6s 6p transition in Cs is compared SM me ¯ 3/2 + kin √2 HLeeR + h.c. to the frequency of→ an optical resonator with an invar L − v 4 2 † † 2 † spacer of length L = 12 cm. The lifetime of the atomic µ H H + λ H H + µφhφH H, (8) −9 − upper state here is τat = 30.5 10 s. The speed of sound 5 · −5 SM for steel is vs 6 10 cm/s, and τcav =2 10 s. If we where kin stands for the SM matter field’s kinetic terms, ≈ · · L 2 assume that all fundamental constants oscillate at some me being the electron mass, µ (λ) being the Higgs common frequency fa, then the experiment [13] is sensi- quadratic (quartic) coupling and m (µφh) are the sin- tive to different combinations of constants depending on glet mass (cubic coupling) and higher order terms be- −1 the frequency fa. If fa τcav then Eq. (6) holds. If ing suppressed. The electromagnetic interactions for the τ −1 f τ −1, then the≪ cavity is sensitive only to the electron field, relevant for low energy physics discussed cav ≪ a ≪ at averaged values of EH and α, while the atoms maintain below are: sensitivity to the variation. As a result, µ gauge =¯eA γ e . (9) L µ δ (νat/νcav) δE = H [1 + (αZ)] , (7) The coupling µφh in Eq. (8) induces mixing between ϕ νat ( /νcav) EH O and h with the mixing angle usually designated as θ (see for instance [28] for a recent review). Then we find that where δEH = EH EH is the deviation from the time the Yukawa interaction in (8) between the electron and − h i averaged value. the field H leads to a similar term between the electron Equation (7) shows that for intermediate frequencies and the scalar field ϕ: νat fa the ratio /νcav depends on the variation of the dimen- ϕ sionful parameter E . At this point we need to specify sin θ √2 me e¯LeR . (10) H v what kind of models we are interested in. Discussion of models. First, we assume that at short At the one-loop level, a coupling between the scalar ϕ distances our system is described by a local perturbative and the photon is induced (see, for example, [29]), ap- Lorentz invariant quantum field theory (QFT), which im- proximately given by: plies no CPT violation. For this case we have fairly good α ϕ µν understanding of how to proceed. Without loss of gener- sin θ Fµν F . (11) 4π v ality, we are allowed to use natural units ~ = c = 1 (see, Of course, there will be similar induced terms for other for instance, [22–24]). We also have examples of working particles and interactions of the standard model, which models (for instance, dilation, relaxion, and SUSY the- we omitted here for simplicity. We note that couplings of ories). For the gauge field we can use a normalization similar form are also expected for a simple dilaton model where the coupling constant α is absorbed into the field which couples to the gauge fields via the anomalous con- (αAµ Aµ)[25]. Then the kinetic term for the gauge → 1 µν tribution to the trace of the energy-momentum tensor field has the form: kin = F F . L − 4α µν (see for instance [5, 30]). Using the above conventions we can now consider a Now we can introduce a time and a space dependent model with relevant fields (omitting for simplicity the classical field ϕ and see what the implications are. This weak and strong gauge fields): Aµ, the photon field (with leads to a theory with a space and time dependent effective Higgs-VEV. The terms in Eqs. (10) and (11) modify the kinetic and mass terms in the Lagrangian of Eq. (8). Implications of these modifications are the same as those 4 The material of the cavity is generally important for precision of varying coupling constant α(ϕ) and mass me(ϕ): measurements. The length of a crystalline cavity is conceptually connected to fundamental constants, which is different from a 1 ϕ cavity based on amorphous glass. However, in the frequency α(ϕ)= α 1 + sin θ , (12a) range considered in this work for fast variations of fundamental π v constants, the relatively slow creeps of the glass material should ϕ not make a significant contribution. me(ϕ)= me 1 sin θ . (12b) − v 4

We see that in this model an effective variation of the Summary. The presence of oscillating background fine-structure constant α and mass me appears, which is fields in a broad class of QFT models, may be inter- linear in the field φ. preted as temporal variations of fundamental constants. Now we need to find out how this affects atomic unit In the case of variation of a constant q, it is possible to EH in Eq. (7). With the chosen units (~ = c = 1), find setups where q is calibrated by its own average value any variation of the atomic unit EH is induced by the q , resulting in a comparison of a dimensionless ratio hq−hi qi variations of α and me. If we rewrite (1) as hqi . In this sense, for the case of rapid variations, it is possible to test variations of dimensionful constants as E = m c2α2, (13) H e well as that of dimensionless ones. To be sensitive to we see that such variations requires two systems, one of which has a faster response (such as an atom) and another is more δE δm δα inertial (such as a cavity). Then the faster-response sys- H = e +2 . (14) EH me α tem tracks instantaneous values of the constants, while the inertial one depends only on their average values. Substituting (12) into (14), we find a (unit independent) result: Acknowledgements. We are grateful to Roee Ozeri, An- δE ϕ 2 drei Derevianko, and Surjeet Rajendran for enlightening H = sin θ 1 , (15) EH − v − π discussions. This work received support from the Rus- sian Science Foundation under Grant No 19-12-00157, which is connected to the experimental observables via the European Research Council (ERC) under the Euro- Eq.(7). pean Unions Horizon 2020 research and innovation pro- The above mechanism can be actually realized in gram (grant agreement No 695405 and 614794), from dilaton-DM theory [5] and in cases where ϕ is an axion- the Simons and Heising-Simons Foundations, Excellence like DM field that is subject to (spontaneous) CP viola- Cluster PRISMA+, ISF, BSF, Minerva, Segre award, the tion, as in the case of relaxion dark matter models [7, 31]. DFG Reinhart Koselleck project and NIST.

[1] A.M. Safronova, D. Budker, D. DeMille, D.F. Jackson [16] M.P. Savedoﬀ. Nature 1956, 178, 688. Kimball, A. Derevianko, and C. W. Clark, Rev. Mod. [17] D.G. Matei, T. Legero, S. H¨afner, C. Grebing, R. Phys. 2018, 90, 025008. Weyrich, W. Zhang, L. Sonderhouse, J.M. Robinson, J. [2] T. Damour, and A.M. Polyakov, Nucl. Phys. B 1994, Ye, F. Riehle, and U. Sterr. Phys. Rev. Lett., 2017,118, 423, 532. 263202. [3] J. D. Barrow, Phys. Rev. D 1999, 59, 043515. [18] W. Zhang, M. Robinson, L. Sonderhouse, E. Oelker, [4] H.B. Sandvik, J.D. Barrow, and J. Magueijo, Phys. Rev. C. Benko, J.L. Hall, T. Legero, D.G. Matei, F. Riehle, Lett. 2002, 88, 031302. U. Sterr, and J. Ye. Phys. Rev. Lett. 2017, 119, 243601. [5] A. Arvanitaki, J. Huang, and K. Van Tilburg. Phys. Rev. [19] V.V. Flambaum. Phys. Rev. Lett. 2006, 97, 092502. D 2015, 91, 015015. [20] P.G. Thirolf, B. Seiferle, and L. von der Wense, J. Phys. [6] Y.V. Stadnik, and V.V. Flambaum, Phys. Rev. Lett. B: At. Mol. Opt. Phys. 2019, 52, 203001. 2015, 115, 201301. [21] A.A. Geraci, C. Bradley, D. Gao, J. Weinstein, [7] A.Banerjee, H. Kim, and G. Perez. Phys. Rev. D. 2019, and A.Derevianko. Phys. Rev. Lett. 2019, 123, 031304. 100, 115026. [22] D. Tong, Part III Cambridge University Mathematics [8] M.G. Kozlov, and D. Budker, Ann. Phys. (Berlin) 2018, Tripos, Michaelmas, 2006. 1800254. [23] D.B. Kaplan. Arxiv 2005, arXiv:nucl-th/0510023. [9] A. Hees, O. Minazzoli, E. Savalle, and Y.V. Stadnik, [24] M.E. Peskin, and D.V. Schroeder, An Introduction To P. Wolf, Phys. Rev. D 2018, 98, 064051. Quantum Field Theory, CRC Press, Boca Raton, 1999. [10] A. Arvanitaki, P.W. Graham, J.M. Hogan, S. Rajendran, [25] F. Wilczek, Quantum Field Theory. In: B. Bederson, and K. Van Tilburg, Phys. Rev. D 2018, 97, 075020. More Things in Heaven and Earth, Springer, New York, [11] E. Savalle, B.M. Roberts, F. Frank, P.E. Pottie, 1999. B.T. McAllister, C. Dailey, A. Derevianko, and P. Wolf, [26] M. Tanabashi, et al. (Particle Data Geoup). Phys. Rev. ArXiv 2019, arXiv:1902.07192. D 2018, 98, 030001. [12] S. Aharony, N. Akerman, R. Ozeri, G. Perez, I. Savoray, [27] European Strategy for Particle Physics Preparatory Group, and R. Shaniv, Arxiv 2019, arXiv:1902.02788 [hep-ph]. Arxiv 2019, arXiv:1910.11775 [hep-ex]. [13] D. Antypas, O. Tretiak, A. Garcon, R. Ozeri, G. Perez, [28] J. Beacham, et al. Arxiv 2019, arXiv:1901.09966 [hep- and D. Budker, Phys. Rev. Lett. 2019, 123, 141102. ex]. [14] A. Arvanitaki, S. Dimopoulos, and K. Van Tilburg, [29] A. Djouadi, Phys. Rept. 2008, 457, 1. Phys. Rev. Lett. 2016, 116, 031102. [30] W.D. Goldberger, B. Grinstein, and W. Skiba, Phys. [15] V.A. Dzuba, V.V. Flambaum, and J.K. Webb, Phys. Rev. Lett. 2008,100, 111802. Rev. Lett. 1999, 82, 888. [31] T. Flacke, C. Frugiuele, E. Fuchs, R.S. Gupta, 5 and G. Perez. J. High Energ. Phys, 2017 06, 50.