[Paper Review] SIDIS transverse spin azimuthal asymmetries at COMPASS: Multidimensional analysis

B. Parsamyan|arXiv (Cornell University)|Dec 21, 2015

Particle physics theoretical and experimental studies22 references3 citations

TL;DR

This paper presents the first multidimensional analysis of transverse spin azimuthal asymmetries in semi-inclusive deep inelastic scattering (SIDIS) using data from the COMPASS experiment at CERN, extracting x-, z-, Q²-, and pT-dependent asymmetries for Sivers, Collins, and other transverse spin distributions. The key result is a direct, unprecedented input for testing the Q² evolution of transverse momentum-dependent (TMD) parton distribution functions, with first evidence of possible Q² dependence in the Sivers effect and non-zero signals for A_UT^{sin(ϕ_S)} and A_LT^{cos(ϕ_h−ϕ_S)}, while other asymmetries are consistent with zero.

ABSTRACT

Exploration of transverse spin structure of the nucleon via study of the spin (in)dependent azimuthal asymmetries in semi-inclusive deep inelastic scattering (SIDIS) and Drell-Yan (DY) reactions is one of the main aspects of the broad physics program of the COMPASS experiment (CERN, Switzerland). In past decade COMPASS has collected a considerable amount of polarized deuteron and proton SIDIS data, while recent 2014 and 2015 runs were dedicated to the Drell-Yan measurements. Results on SIDIS azimuthal effects provided so far by COMPASS play an important role in general understanding of the three-dimensional nature of the nucleon. Giving access to the entire "twist-2" set of transverse momentum dependent (TMD) parton distribution functions (PDFs) and fragmentation functions (FFs) COMPASS data are being widely used in phenomenological analyses and experimental data fits. Recent unique and first ever x-$Q^{2}$-z-pT multidimensional results for transverse spin asymmetries obtained by COMPASS serve as a direct and unprecedented input for one of the hottest topics in the field of spin-physics: the TMD $Q^{2}$-evolution related studies. In addition, extraction of the Sivers and all other azimuthal effects from first ever polarized Drell-Yan data collected recently by COMPASS will reveal another side of the spin-puzzle clarifying the link between SIDIS and Drell-Yan branches. This will be a unique possibility to test predicted universality and key-features of TMD PDFs using essentially the same experimental setup and exploring the same kinematical domain. In this review main focus will be given to the very recent results from COMPASS multi-dimensional analysis of transverse spin asymmetries and to the physics aspects of COMPASS polarized Drell-Yan program.

Motivation & Objective

• To perform a comprehensive, multidimensional analysis of transverse spin asymmetries in SIDIS using high-precision COMPASS data.
• To extract x-, z-, Q²-, and pT-dependent distributions of Sivers, Collins, and other transverse spin effects for the first time in a single experiment.
• To provide direct input for testing the Q² evolution of transverse momentum-dependent (TMD) parton distribution functions (PDFs).
• To enable a direct comparison between SIDIS and Drell-Yan processes via the same experimental setup, testing universality of TMD PDFs.
• To probe the sign change of the Sivers function and validate key features of TMD evolution through first-ever polarized Drell-Yan data analysis.

Proposed method

• A multidimensional unfolding technique was applied to extract asymmetries in four-dimensional phase space: x, z, Q², and pT.
• The analysis used the full set of COMPASS proton and deuteron SIDIS data collected in polarized runs, with transverse target polarization measured in the lab frame.
• Theoretical cross-section expressions were used to model azimuthal distributions, incorporating all possible spin-dependent terms up to twist-2, including Sivers, Collins, and TMD fragmentation effects.
• Asymmetries were extracted in 3D and 4D grids, with special attention to Q² dependence for Sivers and Collins effects.
• The analysis included the first extraction of A_UT^{sin(ϕ_S)} and A_LT^{cos(ϕ_h−ϕ_S)} in 3D configurations, with comparisons to theoretical predictions.
• Future comparison with first-ever polarized Drell-Yan data from COMPASS will test universality of TMD PDFs.

Experimental results

Research questions

• RQ1 What is the Q² dependence of the Sivers and Collins transverse spin asymmetries in SIDIS, and does it match theoretical expectations for TMD evolution?
• RQ2 Are the Sivers and Collins effects observed to be non-zero across different x, z, and pT bins, and how do their magnitudes vary?
• RQ3 Is the A_UT^{sin(ϕ_S)} asymmetry non-zero, and can it be associated with Sivers and Collins mechanisms via the Wandzura-Wilczek approximation?
• RQ4 Is the A_LT^{cos(ϕ_h−ϕ_S)} asymmetry the only significant double-spin asymmetry observed, and does it agree with theoretical predictions?
• RQ5 How do the SIDIS results compare with the upcoming polarized Drell-Yan data from the same experiment to test universality of TMD PDFs?

Key findings

• The Sivers asymmetry shows a sizable signal across the entire x-range for positive hadrons, with hints of a decreasing Q² dependence at large z, particularly in high-x bins.
• For negative hadrons, the Sivers signal is less clear but shows possible negative effects at low x and positive at large x and Q², suggesting a complex x- and Q²-dependent structure.
• The Collins asymmetry exhibits a clear mirrored behavior between positive and negative hadrons and increases in magnitude with z and pT, but shows no clear Q² dependence.
• The A_UT^{sin(ϕ_S)} asymmetry is non-zero and sizable at large z for both positive and negative hadrons, indicating a non-trivial spin-orbit correlation.
• The A_LT^{cos(ϕ_h−ϕ_S)} asymmetry is the only double-spin asymmetry found to be statistically significant, with good agreement between data and theoretical predictions from [18].
• The remaining four asymmetries (A_UT^{sin(3ϕ_h−ϕ_S)}, A_UT^{sin(ϕ_h−ϕ_S)}, A_UT^{sin(2ϕ_h−ϕ_S)}, A_UT^{sin(ϕ_h+ϕ_S)}) are consistent with zero within statistical uncertainties, in line with theoretical expectations.