hep-ph0605266/all.tex
1: \documentclass[superscriptaddress,showkeys,showpacs]{revtex4}
2: \usepackage{graphicx}
3: \usepackage[tbtags]{amsmath}
4: 
5: \begin{document}
6: \title{Constraints on gluon polarization in the nucleon at NLO accuracy.}
7: \thanks{Partially supported by CONICET, Fundaci\'on Antorchas, UBACYT and 
8: ANPCyT, Argentina.}
9: \author{G. A. Navarro} 
10: \email{gabin@df.uba.ar}
11: \affiliation{Departamento de F\'{\i}sica,
12: Universidad de Buenos Aires\\ Ciudad Universitaria, Pab.1 (1428)
13: Buenos Aires, Argentina}
14: \author{R. Sassot}
15: \email{sassot@df.uba.ar}
16: \affiliation{Departamento de F\'{\i}sica,
17: Universidad de Buenos Aires\\ Ciudad Universitaria, Pab.1 (1428)
18: Buenos Aires, Argentina}
19: \date{{\bf \today}}
20: 
21: \begin{abstract}
22: We compare constraints on the gluon polarization in the nucleon 
23: obtained in next to leading order global QCD fits to polarized deep inelastic 
24: scattering data with those coming from observables more directly linked to 
25: the gluon polarization, such as the double spin asymmetry measured by 
26: {\sc Phenix} at RHIC, and high-$p_T$ hadron production studied by COMPASS 
27: at CERN.   
28: \end{abstract}
29: 
30: \pacs{13.60.Hb, 13.88.+e}
31: \keywords{Polarized DIS; gluon polarization.}
32: 
33: \maketitle
34: 
35: \section{Introduction}
36: The extent to which gluons are polarized in the nucleon, and consequently the
37: origin of the nucleon spin, has persisted as an elusive question for almost 
38: two decades in spite of strenuous experimental efforts and theoretical 
39: activity \cite{reviews}. 
40: Although the spin dependent gluon density in principle can be sized in 
41: inclusive deep inelastic scattering measurements, mainly through the scale 
42: dependence of the measured asymmetries, this dependence is rather mild 
43: in the kinematical range accessed by experiments, and conclusions about it 
44: are also veiled by our ignorance regarding the polarization of the other 
45: partonic species, which also contribute to the scale dependence, specially 
46: that of sea quarks. Therefore, even in the most ambitious scenario, inclusive 
47: deep inelastic scattering data can at most suggest mild constraints on the 
48: gluon polarization.
49: 
50: In a recent article \cite{deFlorian:2005mw}, we have shown that the enduring
51: efforts to measure less inclusive observables in deep inelastic scattering 
52: have finally begun to yield, allowing combined next to leading order global 
53: QCD fits
54: to inclusive and semi-inclusive deep inelastic scattering data where sea quarks
55: and gluons are much more definitely constrained. In the mean time, independent 
56: measurements of other less inclusive observables, such as pion production in 
57: polarized proton-proton collisions \cite{panic}, and high transverse momentum 
58: hadron pair production in deep inelastic scattering have begun to provide more 
59: direct assessments of the gluon polarization with competing precision 
60: \cite{Ageev:2005pq,Procureur:2006sg}. It is therefore of great interest to compare the gluon 
61: polarization estimates coming from both the global analysis of deep inelastic 
62: scattering data and from the more direct measurements. In the following we 
63: perform such comparison and 
64: we find that although preliminary direct measurements still have a moderate
65: impact in the fits, there is a remarkable agreement and complementarity 
66: between both approaches, what encourage us to incorporate the forthcoming data 
67: in future global analysis.
68: 
69: \section{Comparison}
70: 
71: In the case of inclusive and semi-inclusive polarized deep inelastic scattering
72: (DIS), the next to leading order (NLO) QCD framework  required  to compute the 
73: respective observables have been available for some time 
74: \cite{inclusiva,deFlorian:1995fd}, and indeed recent global analyses have demonstrated both the relevance of 
75: these corrections and also the non negligible impact of most recent 
76: semi-inclusive data \cite{deFlorian:2005mw, deFlorian:2000bm}. Specifically, 
77: in reference \cite{deFlorian:2005mw} it has been found that the best global 
78: QCD fits to combined DIS data constrain the gluon polarization to be 
79: moderately positive with a first moment of this density $\delta g \equiv \int
80: dx \,\Delta g$ of $0.680$ at $10\,GeV^2$ with an uncertainty range given 
81: by $[0.452,0.771]$ for a one-unit increase in $\chi^2$ and by 
82: $[-0.107,0.807]$ allowing a more conservative $2\%$ variation of $\chi^2$.
83: In these constraints, both the requirement of positivity of the polarized 
84: parton densities relative to a modern set of unpolarized parton densities 
85: \cite{MRST02}, and the correlation between gluon and sea quark polarization,
86:  are found to be crucial. Fits with a wide variation in the gluon 
87: polarization reproduce inclusive data equally well, however they are clearly 
88: differentiated because of their sea quark polarization by semi-inclusive data.
89:     
90: The cross section for single inclusive large $p_T$ pion production in 
91: longitudinally polarized proton-proton collisions, which is right now being 
92: measured at Brookhaven National Laboratory Relativistiv Heavy Ion Collider 
93: (BNL RHIC) \cite{Adler:2004ps} have also been computed at NLO 
94: accuracy, and have been found to be significantly dependent on $\Delta g$ 
95: \cite{Jager:2002xm,deFlorian:2002az}.
96: Recently, the {\sc Phenix} collaboration has presented preliminary results with considerably 
97: reduced errors \cite{panic} which clearly disfavors scenarios with large gluon 
98: polarization and are in nice agreement with estimates of updated polarized 
99: fits. In Figure~\ref{deltag}a  we show the expectation for the
100: double spin asymmetry computed with the best fit of reference 
101: \cite{deFlorian:2005mw}, together with the data reported by {\sc Phenix} \cite{panic}. 
102: We also plot the uncertainty band associated to a $\Delta \chi^2= 2\%$ 
103: variation.
104: \setlength{\unitlength}{1.mm}
105: \begin{figure}[hbt]
106: \includegraphics[width=3.5in]{asynew.eps}
107: \includegraphics[width=3.5in]{deltag.eps}
108: \caption{ (a) Data on $A^{\pi^0}_{LL}$ \cite{panic} compared with the estimate 
109: coming from the the NLO fit of reference \cite{deFlorian:2005mw} and the uncertainty band allowing $\Delta \chi^2= 2\%$; (b) the same but for estimates for $\Delta g/g$ from \cite{Procureur:2006sg} .
110: \label{deltag}}
111: \end{figure}
112: The $\chi^2$ for {\sc Phenix} data obtained with the best fit of reference 
113: \cite{deFlorian:2005mw}, which is previous to the latest set of {\sc Phenix}
114:  data, 
115: and computed including the 40\% scaling uncertainty in the nondiagonal 
116: covariance  matrix, results to be $10.93$, for $N=8$ data points, which is 
117: well within the 
118: $\sqrt{2N}$ range expected for a given subset of data in global fit and 
119: therefore should be considered as consistent. 
120: The value for $\chi^2$ is comparable to the one obtained ($11.2$) in a recent 
121: fit \cite{Hirai:2006sr} to both inclusive DIS data and the {\sc Phenix} 
122: measurement, although not including the scaling error 
123: in the computation. 
124: 
125: Similar agreement is found comparing the expectation of the fit 
126: for $\Delta g /g$ at $1$ GeV$^2$ against preliminary data from COMPASS
127: \cite{Ageev:2005pq,Procureur:2006sg}, and previous measurements 
128: \cite{Airapetian:1999ib,Adeva:2004dh}, as shown in Figure~\ref{deltag}b. 
129: In this case we include both the leading order (LO) and the NLO expectation 
130: because the reported values for $\Delta g /g$ correspond to a LO extraction 
131: improved with Montecarlo higher order corrections. 
132: We include in the plot the uncertainty band coming from a 2\% variation  in 
133: $\chi^2$, plus that coming from varying $Q^2$ up to $10$ GeV$^2$, what again
134: highlights the nice consistency between independent data set and the frameworks
135: implemented for the corresponding analyses. 
136: 
137: \section{Combined fit}.
138: 
139: Further insight on the interplay between DIS data and that coming from 
140: {\sc Phenix}
141: can be obtained analyzing the profile of $\chi^2$ function for the 
142: different subsets of data in a combined fit, against the range of variation of 
143: the net gluon polarization, as it was done in \cite{deFlorian:2005mw}. 
144: In Figure~\ref{profile} we show  the profile of the 
145: total $\chi^2${\scriptsize{$(DIS+A^{\pi^0}_{LL})$}} of a global fit to 
146: inclusive and semi-inclusive data along the lines of that of reference 
147: \cite{deFlorian:2005mw} but also including {\sc Phenix}  data as a solid line. 
148: The curve has similar shape to the one found in 
149: \cite{deFlorian:2005mw} but shifted upwards between eleven and thirteen units,
150: which is essentially the partial contribution of $A^{\pi^0}_{LL}$ to 
151: $\chi^2${\scriptsize{$(DIS+A^{\pi^0}_{LL})$}}.
152: In order to see the relevance of $A^{\pi^0}_{LL}$ data, this last contribution 
153: is also plotted as a dashed-dotted line with an offset 430.91 units, which the 
154: partial contribution  of DIS data  $\chi^2_0${\scriptsize$(DIS)$} for the best 
155: fit. 
156: Notice the partial contribution of $A^{\pi^0}_{LL}$ is almost flat around 
157: the minimum of $\chi^2${\scriptsize{$(DIS+A^{\pi^0}_{LL})$}} and reaches its 
158: own minimum for slightly 
159: lower values of $\delta g$ but within the one-unit variation of 
160: $\chi^2${\scriptsize{$(DIS+A^{\pi^0}_{LL})$}}, what highlights the 
161: consistency between both data sets.
162: 
163: \setlength{\unitlength}{1.mm}
164: \begin{figure}[hbt]
165: \includegraphics[width=4.5in]{parabolag.eps}
166: \caption{Profiles of contributions to $\chi^2$ of different sets of data 
167: against the gluon net polarization. 
168: \label{profile}}
169: \end{figure}
170: In  Figure~\ref{profile} we have plotted also the profile obtained 
171: for DIS data only in \cite{deFlorian:2005mw} as a dashed line, 
172: with an offset given by the partial contribution of $A^{\pi^0}_{LL}$ 
173: at its minimun  $\chi^2_0${\scriptsize$(A^{\pi^0}_{LL})$}, in order to see the 
174: net effects on the fit of {\sc Phenix} data. In Table  \ref{tab:table1} we 
175: show the partial contributions to $\chi^2$ at both minima.
176: 
177: \begin{table}[hbt]
178: \begin{tabular}{ccccc} \hline \hline
179:   & $\chi ^2${\scriptsize$(DIS+A^{\pi^0}_{LL})$} &$\chi^2${\scriptsize$(DIS)$} & $\chi^2${\scriptsize$A^{\pi^0}_{LL})$}  & $\delta g$ \\ \hline
180: $\chi ^2_0${\scriptsize$(DIS+A^{\pi^0}_{LL})$}& 441.84&430.91&10.93&0.680 \\ 
181: $\chi ^2_0${\scriptsize$(A^{\pi^0}_{LL})$}& 442.63&431.82&10.81&0.450 \\ \hline \hline
182: \end{tabular}
183: \caption{\label{tab:table1} Partial contributions to $\chi^2$ values and 
184: first moment of $\Delta g$ at $Q^2=10$ GeV$^2$}
185: \end{table}
186: 
187: 
188: Clearly, these effects are almost imperceptible around the minimum but can 
189: be noticed for  $\chi^2${\scriptsize$(DIS+A^{\pi^0}_{LL})$} variations between 
190: the one-unit and the $2\%$ 
191: increase. Close to the minimum of 
192: $\chi^2${\scriptsize{$(DIS+A^{\pi^0}_{LL})$}}, the decrease of  
193:  $\chi^2${\scriptsize$(A^{\pi^0}_{LL})$} for decreasing $\delta g$ is 
194: overpowered by the increase of $\chi^2${\scriptsize$(DIS)$} and consequently,
195: the position of the minimum remains that found for DIS data.
196: Given the large number of DIS data included in the fit (478) compared to the
197: rather limited set of $A^{\pi^0}_{LL}$ available at present, the small impact
198: in the fit is not surprising, nevertheless the consistency shown, and the 
199: possibility of increasing considerably the statistics in the future is 
200: encouraging.     
201: 
202:  
203: \section{Conclusions}
204: We have compared the constraints on the gluon polarization in the nucleon 
205: obtained in next to leading order global QCD fits to polarized deep inelastic 
206: scattering data with those coming  the double spin asymmetry measured by 
207: {\sc Phenix} at RHIC. 
208: Although the relative statistical weight of  $A^{\pi^0}_{LL}$ data in a 
209: global NLO fit including also DIS data is rather limited, we find a remarkable 
210: agreement and moderate improvement when combining both data sets. 
211: In the case of other direct measurements such as high-$p_T$ hadron production 
212: studied by COMPASS at CERN,  the lack of a NLO framework for the computation 
213: of the corresponding asymmetries does not allow to include them yet in a 
214: combined NLO global analysis however we find preliminary agreement which 
215: hopefully will be checked at NLO accuracy in the near future.
216: 
217: 
218: \section{Acknowledgements}
219: 
220: We warmly acknowledge Daniel de Florian for comments and suggestions 
221: and  Y. Fukao for discussions relative to {\sc Phenix} data.
222: 
223: 
224: 
225: \begin{thebibliography}{3}
226: 
227: \bibitem{reviews} B. Lampe and E. Reya, Phys. Rep. 332, 1 (2000);\\
228:  E. Leader et al., Phys, Rep. 261, 1 (1995). 
229: \bibitem{deFlorian:2005mw}
230:  D.~de Florian, G.~A.~Navarro and R.~Sassot,
231:  \emph{Phys.\ Rev.\ D} \textbf{71}, 094018 (2005).
232: \bibitem{panic}{\sc Phenix} collaboration, K. Boyle, talk given at the XVIIth Particles and Nuclei International Conference (Panic05), http://www.panic05.lanl.gov/
233: \bibitem{Ageev:2005pq}
234:   E.~S.~Ageev {\it et al.},
235:   Phys.\ Lett.\ B {\bf 633}, 25 (2006).
236: \bibitem{Procureur:2006sg}
237:   S.~Procureur,
238:   %``New measurement of Delta(G)/G at COMPASS,''
239:   arXiv:hep-ex/0605043.
240:   %%CITATION = HEP-EX 0605043;%%
241: 
242: \bibitem{Airapetian:1999ib}
243:   A.~Airapetian {\it et al.},
244:  Phys.\ Rev.\ Lett.\  {\bf 84}, 2584 (2000).
245: \bibitem{Adeva:2004dh}
246:   B.~Adeva {\it et al.}, Phys.\ Rev.\ D {\bf 70}, 012002 (2004).
247: \bibitem{inclusiva} R. Mertig, W. L. van Neerven, Z.Phys.{\bf C70}, 637,(1996);
248:  W. Vogelsang, Phys.Rev.{\bf D54}, 2023, (1996).
249: \bibitem{deFlorian:1995fd}
250:   D.~de Florian, C.~A.~Garcia Canal and R.~Sassot,
251:   %``Factorization in Semi-Inclusive Polarized Deep Inelastic Scattering,''
252:   Nucl.\ Phys.\ B {\bf 470}, 195 (1996) [arXiv:hep-ph/9510262].
253:   %%CITATION = HEP-PH 9510262;%%
254: \bibitem{deFlorian:2000bm}
255:   D.~de Florian and R.~Sassot,
256:   Phys.\ Rev.\ D {\bf 62}, 094025 (2000).
257: \bibitem{Jager:2002xm}
258:   B.~J\"ager, A.~Schafer, M.~Stratmann and W.~Vogelsang,
259:   %``Next-to-leading order QCD corrections to high-p(T) pion production in
260:   %longitudinally polarized p p collisions,''
261:   Phys.\ Rev.\ D {\bf 67}, 054005 (2003)
262:   [arXiv:hep-ph/0211007].
263:   %%CITATION = HEP-PH 0211007;%%
264: \bibitem{deFlorian:2002az}
265:   D.~de Florian,
266:   %``Next-to-leading order QCD corrections to one hadron production in 
267:   % polarized p p collisions at RHIC,''
268:   Phys.\ Rev.\ D {\bf 67}, 054004 (2003)
269:   [arXiv:hep-ph/0210442].
270:   %%CITATION = HEP-PH 0210442;%%
271: \bibitem{MRST02} A. D. Martin et al., 
272: %R. G. Roberts, W. J. Stirling and R. S. Thorne,
273:  \emph{Eur.\ Phys.\ J. C} \textbf{28} 455 (2003).
274: \bibitem{Adler:2004ps}
275:   S.~S.~Adler {\it et al.}  [{\sc Phenix} Collaboration],
276:   %``Double helicity asymmetry in inclusive mid-rapidity pi0 production for
277:   %polarized p + p collisions at s**(1/2) = 200-GeV,''
278:   Phys.\ Rev.\ Lett.\  {\bf 93} (2004) 202002
279:   [arXiv:hep-ex/0404027].
280:   %%CITATION = HEP-EX 0404027;%%
281: \bibitem{Hirai:2006sr}
282:   M.~Hirai, S.~Kumano and N.~Saito,
283:   %``Determination of polarized parton distribution functions with recent data
284:   %on polarization asymmetries,''
285:   arXiv:hep-ph/0603213.
286:   %%CITATION = HEP-PH 0603213;%%
287: \end{thebibliography}
288: 
289: 
290: 
291: \end{document} 
292: 
293: