Drell-Yan Angular Analysis

• The paper extracts angular coefficients (A0, A2, etc.) using a reweighted maximum-likelihood fit to rigorously test QCD predictions and the Lam–Tung relation.
• Drell–Yan angular analysis is a quantitative study of lepton-pair distributions that reveals key aspects of partonic dynamics and gluon spin behavior.
• The measurement confirms the Lam–Tung relation, supporting the vector nature of gluons and constraining models of QCD spin-correlation and parton-level processes.

The Drell-Yan angular analysis is the quantitative study of angular distributions of lepton pairs produced via the annihilation of quark and antiquark in hadron–hadron collisions (specifically $p\bar{p}\to\gamma^*/Z\to e^+e^-+X$ at $\sqrt{s}=1.96$ TeV, as executed by the CDF II experiment). It provides a direct probe of the partonic-level dynamics, the vector nature of the gluon, QCD radiative effects, and spin–correlation phenomena, by extracting a set of angular coefficients $A_i$ from the observed lepton angular spectrum in the Collins–Soper frame. The complete measurement disentangles the polarization states and their modulations in transverse momentum $P_T$, enabling a rigorous test of the Lam–Tung relation and the underlying theory.

1. Collins–Soper Angular Decomposition and Coefficient Definitions

In the Collins–Soper rest frame, the polar angle $\theta$ and azimuthal angle $\phi$ of the $e^+$ (or $e^-$) are defined with respect to the bisector of the beam directions. The differential cross section is expressed as: $$\frac{d\sigma}{d\Omega} \propto (1+\cos^2\theta) + \frac{A_0}{2}(1-3\cos^2\theta) + A_1\sin2\theta\cos\phi + \frac{A_2}{2}\sin^2\theta\cos2\phi + A_3\sin\theta\cos\phi + A_4\cos\theta + A_5\sin^2\theta\sin2\phi + A_6\sin2\theta\sin\phi + A_7\sin\theta\sin\phi$$ Each coefficient $A_i$ parameterizes a specific physical correlation:

• $A_0$: Longitudinal versus transverse polarization admixture of $\gamma^*/Z$
• $A_2$: $\cos2\phi$ azimuthal modulation from spin–spin and orbital angular momentum
• $A_1$: $\sin2\theta \cos\phi$ term (spin–flip interference)
• $A_3$, $A_4$: Parity-violating terms from $\gamma^*$–$Z$ interference (forward–backward and $\cos\phi$ asymmetry)
• $A_5$, $A_6$, $A_7$: T-odd/higher-twist terms (predicted to vanish at leading order in perturbative QCD)

The extraction of these coefficients provides a minimal set necessary for full angular characterization.

2. Experimental Realization: CDF II Detector and Event Selection

CDF II employed a multi-layer tracking system (SVX, COT) in a solenoidal magnetic field, projective-tower EM/HAD calorimetry, and high-$E_T$ triggers. Drell–Yan events were selected in the $Z$ mass window $66<M_{ee}<116$ GeV/c$^2$ and divided into CC, CP, and PP topologies according to electron rapidities and tracker associations. Electrons were required to exceed $E_T$ thresholds and be matched to SVX tracks; QCD and electroweak backgrounds were carefully subtracted (total background $<0.5\%$). The dataset comprised $\sim$140,000 events over $2.1$ fb$^{-1}$ of integrated luminosity.

3. Angular Coefficient Extraction and Systematics

The analysis binned events in five $P_T$ intervals ($\langle P_T\rangle$ from $4.8$ to $73.7$ GeV/c):

• $[0-10]$, $[10-20]$, $[20-35]$, $[35-55]$, $[>55]$ GeV A re-weighted maximum-likelihood fit was performed on the $(\cos\theta,\,\phi)$ distributions using PYTHIA+GEANT simulated events. Each $P_T$ bin’s data was fitted for $A_0$ and $A_4$ (from $\cos\theta$), and $A_2$, $A_3$ (from $\phi$), floating normalizations. Acceptance, efficiency, resolution, and remaining backgrounds were fully modeled. Systematic uncertainties (electron ID/tracking, background subtraction, material) were propagated through repeated fits, found to be subdominant to statistical errors.

Measured coefficients ($A_0$, $A_2$, $A_3$, $A_4$ in units of $10^{-1}$):

$P_T$ bin (GeV)	$\langle P_T\rangle$	$A_0$	$A_2$	$A_3$	$A_4$
0–10	4.8	0.17±0.14±0.07	0.16±0.26±0.06	-0.04±0.12±0.01	1.10±0.10±0.01
10–20	14.1	0.42±0.25±0.07	-0.01±0.35±0.16	0.18±0.16±0.01	1.01±0.17±0.01
20–35	26.0	0.86±0.39±0.08	0.52±0.51±0.29	0.14±0.24±0.01	1.56±0.26±0.01
35–55	42.9	3.11±0.59±0.10	2.88±0.84±0.19	-0.19±0.41±0.04	0.52±0.42±0.03
>55	73.7	4.97±0.61±0.10	4.83±1.24±0.02	-0.47±0.56±0.02	0.85±0.50±0.05

4. Lam–Tung Relation and Validation

The Lam–Tung relation, $A_0=A_2$, is a consequence of the spin-1 nature of gluon emission in both $q\bar{q}\to\gamma^*/Zg$ and $qg\to\gamma^*/Zq$ subprocesses. Experimental data yield

$A_0-A_2$ = $\{0.00\pm0.03,$ $0.04\pm0.05,$ $0.03\pm0.07,$ $0.02\pm0.11,$ $0.01\pm0.14\}$ across $P_T$ bins, with average $\langle A_0-A_2\rangle=0.02\pm0.02$. This result is robustly consistent with zero, confirming the Lam–Tung relation to within $5\%$ at high $P_T$, and directly verifying that the gluon behaves as a vector boson in hadronic collisions.

5. Comparison with Theory and QCD Mechanisms

The measured $A_0$ and $A_2$ at low $P_T$ conform to the annihilation formula: $A_0^{q\bar q}=A_2^{q\bar q}=P_T^2/(M_Z^2+P_T^2)$, while at high $P_T$ they exceed this, evidencing $qg$ Compton scattering contributions. Comparison with predictions from NLO generators (DYRAD, POWHEG, FEWZ, MG+PS), matched parton shower models (PYTHIA+1jet), and NNLO corrections indicate excellent agreement with the data, whereas LO-only implementations and resummation-only generators (VBP) underestimate $A_0$ and $A_2$ at large $P_T$.

6. Implications for Gluon Spin, QCD Spin-Correlation, and Global Fits

The precise verification of $A_0=A_2$ rules out a sizable scalar-gluon component, confirming the vector nature of the gluon. The $P_T$ dependence of $A_0$ and $A_2$ quantitatively demonstrates the role and importance of quark–gluon Compton scattering ($qg\to\gamma^*/Z\,q$), as predicted by pQCD at Tevatron energies. The measurement constrains parton-level angular correlations employed in global QCD fits and bolsters the validation of Monte Carlo tools incorporating NLO+parton shower matching. Future analyses at the LHC and other higher-energy experiments will further elucidate the balance of $q\bar{q}$ vs. $qg$ initial-state contributions and probe possible small-$x$ effects or higher-twist corrections in the deviation from leading-twist behavior.

7. Extended Applications and Outlook

The comprehensive approach employed in the CDF angular analysis is extensible to higher energies and alternative kinematic regimes. Measurements of all leading Drell–Yan angular coefficients at high mass and $P_T$ provide critical inputs for the study of QCD spin correlations, the development of resummation techniques, and the extraction of electroweak parameters. The established agreement between data and theory signifies that perturbative QCD, supplemented by sophisticated simulation and fit techniques, remains a powerful framework for describing and predicting spin-dependent structures in neutral-current Drell–Yan processes (Collaboration et al., 2011).