\Upsilon(\mathrm{nS}) polarizations in \mathrm{p} \mathrm{p} collisions at \sqrt{s} = 7 and 8\mathrm{\,Te\kern-1.0ptV} by the LHCb collaboration

polarizations in collisions at  = 7 and 8 by the LHCb collaboration


A polarization measurement carried out for the , and  mesons produced in pp collisions at and 8 is presented. Data samples used for the polarization measurement were collected by the LHCb experiment during the 2011 and 2012 data taking runs with integrated luminosities of 1 and 2, respectively. The measurement has been performed in three polarization frames, using an angular distribution analysis of the  decays in the kinematic region of the transverse momentum and rapidity . No large polarization is observed.

1 Introduction

It is already forty years since the discovery of the first bottomonium state, the  meson [1], but studies of heavy quarkonium production continue to play an important role in the development of quantum chromodynamics (QCD) [2]. According to the current theoretical framework, nonrelativistic QCD (NRQCD) [3, 4], inclusive production of a heavy quarkonium is viewed as a two-step process. In the first step, a heavy quark-antiquark pair, , is perturbatively created in a color singlet or color octet state, and then, in the second step, the hadronization process non-perturbatively transforms the pair into an observable colorless bound state. The non-perturbative transitions are described by long-distance matrix elements which are conjectured to be independent of production processes in the first step, and need to be extracted experimentally. The most distinct signature of the first NRQCD calculations [5, 6, 7, 8] was the prediction of transverse polarization for S-wave quarkonium states (such as the , and mesons) directly produced (i.e. not coming from decays) at large transverse momentum in high-energy hadron collisions. NRQCD calculations for promptly produced heavy quarkonia (if charmonia then those not coming from b-hadron decays) are complicated by feed-down (electromagnetic or hadronic transitions) from higher level states.

Full NLO calculations [9] performed for the , and mesons (shortly denoted by ), including effects of feed-down contributions for the and but not for the , give very small transverse polarizations for the first two bottomonium states and a large transverse polarization for the third state. These calculations were done before LHCb obtained that the fractions of mesons originating from  decays are around for high transverse momenta,  [10]. The was considered as almost feed-down free state before. Although taking into account feed-down contributions improves the description of polarization, there are still problems in describing the polarization [11] (even with considering feed-down) and the polarization [12] (which includes negligible feed-down contributions).

The angular distribution of muons from the  decay can be written as [13, 14, 15]


where the angular quantities describe a direction of in the rest frame with respect to some specified axes, are the angular distribution parameters directly related to the spin-1 density-matrix elements [13, 16, 17]. The parameter is a measure of spin-alignment, and can be expressed as , where () is the transverse (longitudinal) component of the cross section. If the spin-alignment parameter (), the meson is called to be transversely (longitudinally) polarized in a specified frame, while the case means that the meson is unpolarized. The parameters depend on a definition of coordinate axes specified in the rest frame. The following three coordinate systems are widely used in polarization analyses: helicity (HX) [18], Collins-Soper (CS) [19] and Gottfried-Jackson (GJ) [19]. The frames are specified by different directions of the spin-quantization axis,  axis, defined in the production plane of meson [15]. In all these frames, the  axis is normal to the production plane [15, 21], and the remaining  axis completes a right-handed coordinate system.

Until recently, quarkonium polarization measurements have been reduced only to studies of the parameter , some times in different polarization frames. As pointed out in [15], measuring all the three polarization parameters is important from the theoretical and experimental points of view. Since from having the in one frame, it is possible to transform them into another [15, 22], and perform some cross checks of results. In particular, an important cross check is provided by a polarization parameter  [23, 24], which is invariant for all rotations around the axis, that is invariant in the HX, CS and GJ frames. The physical meaning of the parameter was first recognized in [25] (see also [26, 27]).

The first full angular distribution analysis of muons from the  decays was performed by the CDF collaboration [28] using data of collisions at . CDF found that the angular distributions of muons from all the three states are nearly isotropic in the central rapidity region and . This result is consistent with the previous CDF measurement [29], and inconsistent with the measurement performed by the D0 collaboration [30]. D0 observed the significant dependent longitudinal polarization for the mesons produced in collisions at , for and . The next full angular distribution analysis for the  was performed by the CMS collaboration [31] using collisions data at , for the rapidity ranges and , and for  [31]. CMS found no evidence of large transverse or longitudinal polarization for any of the three mesons in the explored kinematic region. The experimental situation is complicated by the result of the fixed-target experiment E866 [32], which performed the polarization measurement of the mesons produced in p-Cu collisions at in and . The E866 collaboration found that the meson is produced weakly polarized, while the and mesons are produced with a maximal transverse polarization. Although different production energies may determine different dominant contributions in the production processes, all these results underscore the need for further experimental study of the polarization.

The LHCb collaboration performed the full angular distribution analysis [21] for the mesons produced in collisions at and 8 in the LHCb setup during the 2011 and 2012 data taking runs with integrated luminosities of 1 and 2, respectively. The polarization measurement was done in the HX, CS and GJ frames in the kinematic range defined by and .

2 LHCb detector and selection of  decays

The LHCb detector is a single-arm forward spectrometer primarily designed to look for indirect evidence of new physics in violation and rare decays of charm and beauty hadrons. The detector covers a unique pseudorapidity range  corresponding to of the solid angle, but of all produced pairs fall into this geometrical acceptance. A detailed description of the LHCb detector and its performance are given in [33, 34].

The selection of candidates is similar to those used in the previous LHCb analyses [35, 36, 37, 38]. The candidates are formed from pairs of oppositely charged tracks reconstructed in the tracking system. Each track is required to have a good reconstruction quality [39] and to be identified as a muon [40]. Each muon is then required to have momentum satisfying , and pseudorapidity within the region . The two muons are required to originate from a common vertex with a good probability of the vertex fit. In addition, the consistency of the dimuon vertex with a primary vertex is ensured via the quality requirement of a global fit, performed for each dimuon candidate using the primary vertex position as a constraint [41]. This global fit requirement also reduces the background caused by genuine muons coming from decays of long-lived charm and beauty hadrons. A large fraction of the combinatorial background is populated at large values of , where is the polar angle of in the GJ frame. To reduce this background, a requirement is applied. Finally, the mass of the muon pair is required to be in the range .

As an example, a dimuon mass distribution of the  candidates finally selected in and for data set is shown in Fig. 1. The dimuon mass distribution is parametrized by the sum of three double-sided Crystal Ball functions [42, 43] for describing the mesons and an exponential function for the combinatorial background. The parametrization is done by an unbinned extended maximum likelihood fit. The dimuon mass fit is performed in each bin, and results of the fit are then used by the sPlot analysis [44] for obtaining so called signal sWeights: , and . The sWeights are assigned to each, , dimuon candidate for extracting the appropriate signal component. The total signal yields obtained from the dimuon mass fit in the full explored range of and are candidates, candidates and candidates in () data set. The average mass resolution of the peak is .

Figure 1: Dimuon mass distribution in the region , for data obtained at . The thick dark yellow solid curve shows the result of the fit, as described in the text. The three peaks, shown with thin red solid lines, correspond to the , and signals (left to right). The background component is indicated with a dashed blue line.

3 Polarization analysis

The polarization measurement is performed using an unbinned maximum likelihood approach [45] already applied in the and polarization analyses [11],[12]. The polarization parameters are determined from fits to the two-dimensional angular distribution of from the  decay, described by Eq. 1. In each bin, the following logarithm of the likelihood function is constructed for each state:


where , is normalization factor determined by Monte Carlo for each meson, is one of the sWeights for the candidate, and is the total number of all selected candidates in a considered bin. The constant scale factor is introduced to take into account correctly the effect of the sWeights on statistical uncertainties of the parameters obtained after the polarization fit. The influence of the was validated by pseudoexperiments. The normalization factor is defined as


and is calculated using simulated events. In the simulation, where the  mesons are generated unpolarized, the two-dimensional angular distribution of from decays of selected candidates is proportional to the total efficiency , so is evaluated by summing over the selected  candidates in the simulated sample


where is a muon identification efficiency measured directly from data using a large sample of low-background  events (no muon identification requirement was applied when selecting the candidates in the simulated samples); is a correction factor for MC, obtained using data-driven techniques to account for small differences between data and simulation in a tracking efficiency of muons [39, 40] and in the and spectra [46, 47].

4 Results and conclusions

Different sources of systematic uncertainty have been considered when determining the polarization parameters, namely systematic uncertainty related to: a) the signal extraction procedure; b) the muon identification efficiency; c) the track reconstruction efficiency; d) a possible small difference in the trigger efficiency between data and simulation; e) correction factors for the muon identification efficiency; f) the finite size of the simulated samples. All these sources have been studied for the polarization parameters , , and in the HX, CS and GJ frames for each bin. It was found that the most dominant systematic uncertainty is related to the finite size of MC samples, varying between and of the statistical uncertainty. The total systematic uncertainty for each polarization parameter is calculated as the quadratic sum of systematic uncertainties coming from all the considered sources, assuming no correlation between them. For some high- bins the systematic and statistical uncertainties are comparable.

All the polarization results obtained by the LHCb collaboration for data collected at and are given in [21]. Here we outline the main features of the results. The values of the parameter obtained for the mesons do not show any significant transverse or longitudinal polarization in all frames over the considered kinematic region. The values of the parameters and are small in all frames over all bins. All the three polarization parameters do not manifest a distinct dependence on the . The values of the frame invariant parameter measured in the HX, CS and GJ frames are consistent with each other. Moreover, all values of the are close to zero in all phase-space bins. In the considered phase space domain, the polarization results corresponding to and 8 are in good agreement with each other.

The parameters have been checked for positivity constraints imposed on the spin-1 density-matrix [15, 21, 48, 26, 27]. The values of the satisfy all the six positivity constraints in all frames over all phase-space bins. In particular, Fig. 2 shows regions allowed by the positivity constraints together with the parameters and measured in all  bins, for data collected at  and . Further, since the axes of the HX, CS and GJ frames coincide in the limit  [15, 22], we checked this constraint and found that all values of the are very similar for low- bins in all frames. It was also found that the parameters and are very close to zero in the limit , in accordance with kinematic constraints pointed out in [14].

Figs. 3 and 4 show a comparison of the LHCb results [21] with results obtained by the CDF [28] and CMS [31] collaborations for the HX and CS frames, respectively. There is good agreement with CMS results for both frames, and with CDF for the CS frame.

The LHCb collaboration continues to perform measurements devoted to -hadron and quarkonium physics [49, 50, 51, 52]. In particular, the polarization studies performed by LHCb [11, 12, 21] allowed one to shed some new facts on production mechanism of heavy quarkonium states, and to aggravate old ones.

Figure 2: The measured values of  for a) , b)  and c)  mesons. The red solid circles, blue open squares and green solid diamonds correspond to the helicity (HX), Collins-Soper (CS) and Gottfried-Jackson (GJ) frames, respectively. The thick black lines show the regions allowed by the positivity constraints.
Figure 3: The values of  (top),  (middle) and  (bottom) parameters, measured in the HX frame for  (left),  (center) and  (right)  mesons. Results of the LHCb analysis for the rapidity region are shown with red solid circles and blue solid squares for data collected in  collisions at  and , respectively. The results by the CMS collaboration [31] obtained in  collision at  for rapidity regions  and  are shown with magenta open upward triangles and cyan open downward triangles, respectively. The results obtained by the CDF collaboration [28] in  collision at  for rapidity region  are shown with green open diamonds. Some data points are displaced from the bin centers to improve visibility. The error bars indicate the sum of the statistical and systematic uncertainties added in quadrature.
Figure 4: The values of  (top),  (middle) and  (bottom) parameters, measured in the CS frame for  (left),  (center) and  (right)  mesons. Results of the LHCb analysis for the rapidity region are shown with red solid circles and blue solid squares for data collected in  collisions at  and , respectively. The results by the CMS collaboration [31] obtained in  collision at  for rapidity regions  and  are shown with magenta open upward triangles and cyan open downward triangles, respectively. The results obtained by the CDF collaboration [28] in  collision at  for rapidity region  are shown with green open diamonds. Some data points are displaced from the bin centers to improve visibility. The error bars indicate the sum of the statistical and systematic uncertainties added in quadrature.

I wish to thank A.V. Efremov and O.V. Teryaev for their invitation to the DSPIN-17 workshop and for their warm hospitality, and also S.R. Slabospitsky and O.V. Teryaev for useful and interesting discussions.



  1. Herb S W et al, 1977 Phys. Rev. Lett. 39 252
  2. Braaten E and Russ J, 2014 Annu. Rev. Nucl. Part. Sci. 64 221 (Preprint arXiv:1401.7352)
  3. Caswell W E and Lepage G P, 1986 Phys. Lett. B 167 437
  4. Bodwin G T, Braaten E and Lepage G P, 1995 Phys. Rev. D 51 1125 (Preprint arXiv:hep-ph/9407339) \nonumBodwin G T, Braaten E and Lepage G P, 1997 Phys. Rev. D 55 5853
  5. Cho P and Wise M B, 1994 Phys. Lett. B 346 129 (Preprint arXiv:hep-ph/9411303)
  6. Beneke M and Rothstein I Z, 1995 Phys. Lett. B 372 157 (Preprint arXiv:hep-ph/9509375)
  7. Beneke M and Krämer M, 1996 Phys. Rev. D 55 R5269 (Preprint arXiv:hep-ph/9611218)
  8. Leibovich A K, 1997 Phys. Rev. D 56 4412 (Preprint arXiv:hep-ph/9610381)
  9. Gong B, Wan L-P, Wang J-X and Zhang H-F, 2014, Phys. Rev. Lett. 112 032001 (Preprint arXiv:1305.0748)
  10. Aaij R et al [LHCb collaboration], 2014 Eur. Phys. J. C 74 3092 (Preprint arXiv:1407.7734)
  11. Aaij R et al [LHCb collaboration], 2013 Eur. Phys. J. C 73 2631 (Preprint arXiv:1307.6379)
  12. Aaij R et al [LHCb collaboration], 2014 Eur. Phys. J. C 74 2872 (Preprint arXiv:1403.1339)
  13. Oakes R J, 1966 Nuovo Cim. A 44 440
  14. Lam C S and Tung W K, 1978 Phys. Rev. D 18 2447
  15. Faccioli P, Lourenço C, Seixas J and Wöhri H K, 2010 Eur. Phys. J. C 69 657 (Preprint arXiv:1006.2738)
  16. Pilkuhn H M, 1979 “Relativistic Particle Physics” Springer-Verlag
  17. Beneke M, Krämer M and Vänttinen M, 1998 Phys. Rev. D 57 4258 (Preprint arXiv:hep-ph/9709376)
  18. Jacob M and Wick G C, 1959 Ann. Phys. 7 404
  19. Collins J C and Soper D E, 1977 Phys. Rev. D 16 2219
  20. Gottfried K and Jackson J D, 1964 Nuovo Cim. 33 309
  21. Aaij R et al [LHCb collaboration], 2017 Preprint arXiv:1709.01301
  22. Falciano S et al, 1986 Z. Phys. C 31 513
  23. Faccioli P, Lourenço C and Seixas J, 2010 Phys. Rev. Lett. 105 061601 (Preprint arXiv:1005.2601)
  24. Faccioli P, Lourenço C and Seixas J, 2010 Phys. Rev. D 81 111502 (Preprint arXiv:1005.2855)
  25. Teryaev O V, 2006 Proceedings of XI Advanced Research Workshop on High Energy Spin Physics, DUBNA-SPIN-05, Dubna, September 27–October 1, 2005 (ed. by Efremov A V and Goloskokov S V)
  26. Teryaev O V, 2011 Nucl. Phys. Proc. Suppl. 214 118
  27. Teryaev O V, 2011 Proceedings of XIV Advanced Research Workshop on High Energy Spin Physics, DUBNA-SPIN-11, Dubna, September 20–24, 2011 (ed. by Efremov A V and Goloskokov S V)
  28. Aaltonen T et al [CDF collaboration], 2012 Phys. Rev. Lett. 108 151802 (Preprint arXiv:1112.1591)
  29. Acosta D et al [CDF collaboration], 2002 Phys. Rev. Lett. 88 161802
  30. Abazov V M et al [D0 collaboration], 2008 Phys. Rev. Lett. 101 182004 (Preprint arXiv:0804.2799)
  31. Chatrchyan S et al [CMS collaboration], 2013 Phys. Rev. Lett. 110 081802 (Preprint arXiv:1209.2922)
  32. Brown C N et al [E866 collaboration], 2001 Phys. Rev. Lett. 86 2529
  33. Alves A A et al [LHCb collaboration], 2008 JINST 3 S08005
  34. Aaij R et al [LHCb collaboration], 2014 Int. J. Mod. Phys. A 30 1530022 (Preprint arXiv:1412.6352)
  35. Aaij R et al [LHCb collaboration], 2012 Eur. Phys. J. C 72 2025 (Preprint arXiv:1202.6579)
  36. Aaij R et al [LHCb collaboration], 2013 JHEP 06 064 (Preprint arXiv:1304.6977)
  37. Aaij R et al [LHCb collaboration], 2014 Eur. Phys. J. C 74 2835 (Preprint arXiv:1402.2539)
  38. Aaij R et al [LHCb collaboration], 2015 JHEP 11 103 (Preprint arXiv:1509.02372)
  39. Aaij R et al [LHCb collaboration], 2014 JINST 10 P02007 (Preprint arXiv:1408.1251)
  40. Aaij R et al [LHCb collaboration], 2013 JINST 8 P10020 (Preprint arXiv:1306.0249)
  41. Hulsbergen W D, 2005 Nucl. Instrum. Meth. A 552 566 (Preprint arXiv:physics/0503191)
  42. Skwarnicki T, 1986 PhD thesis, Institute of Nuclear Physics, Krakow (DESY-F31-86-02)
  43. Aaij R et al [LHCb collaboration], 2011 Phys. Lett. B 707 52 (Preprint arXiv:1109.0963)
  44. Pivk M and Le Diberder F R, 2004 Nucl. Instrum. Meth. A 555 356 (Preprint arXiv:physics/0402083)
  45. Xie Y, 2009 Preprint arXiv:0905.0724
  46. Sjöstrand T, Mrenna S and Skands P, 2006 JHEP 05 026 (Preprint arXiv:hep-ph/0603175)
  47. Belyaev I et al, 2011 J. Phys. Conf. Ser. 331 032047
  48. Palestini S, 2010 Phys. Rev. D 83 031503 (Preprint arXiv:1012.2485)
  49. Altarelli M P, 2013 EPJ Conf 60 15005 (Preprint arXiv:1307.1110)
  50. Manca G, 2014 Int. J. Mod. Phys. A 29 1430014
  51. Yang Z, 2014 Nucl. Phys. A 931 643
  52. Belyaev I M and Egorychev V Yu, 2015 Phys. Atom. Nucl. 78 977
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description