1: \documentclass[12pt,epsfig]{article}

2: 

3: \textwidth=160mm

4: \textheight=240mm

5: 

6: \setlength{\voffset}{-30mm}

7: \setlength{\hoffset}{-5mm}

8: 

9: \usepackage{graphicx,color,here}

10: \usepackage{here}

11: 

12: \newcommand{\be}{\begin{equation}}

13: \newcommand{\ee}{\end{equation}}

14: \newcommand{\bea}{\begin{eqnarray}}

15: \newcommand{\eea}{\end{eqnarray}}

16: \newcommand{\PR}[1]{{\it Phys.\ Rev.}\ {\bf #1}}

17: \newcommand{\PRL}[1]{{\it Phys.\ Rev.\ Lett.}\ {\bf #1}}

18: \newcommand{\PL}[1]{{\it Phys.\ Lett.}\ {\bf #1}}

19: \newcommand{\zh}{z_h}

20: \newcommand{\bfk}{\mbox{\boldmath $k$}}

21: \newcommand{\bfq}{\mbox{\boldmath $q$}}

22: \newcommand{\bfP}{\mbox{\boldmath $P$}}

23: \newcommand{\bfS}{{\mbox{\boldmath $S$}}_{_T}}

24: \newcommand{\bfl}{\mbox{\boldmath $\ell$}}

25: \newcommand{\pup}{p^\uparrow}

26: \newcommand{\pdown}{p^\downarrow}

27: \newcommand{\qup}{q^\uparrow}

28: \newcommand{\qdown}{q^\downarrow}

29: \def\slash{\rlap{/}}

30: \newcommand{\bfp}{\mbox{\boldmath $p$}}

31: \newcommand{\Lup}{\Lambda^\uparrow}

32: \newcommand{\Ldown}{\Lambda^\downarrow}

33: \newcommand{\Aup}{A^\uparrow}

34: \newcommand{\hup}{h^\uparrow}

35: \newcommand{\hdown}{h^\downarrow}

36: \def\bkt{\bf k_\perp}

37: \def\bpt{\bf p_\perp}

38: \def\kt{k_\perp}

39: \def\pt{p_\perp}

40: 

41: %

42: \begin{document}

43: 

44: \title{\bf Cahn and Sivers effects in the target fragmentation region of SIDIS}

45: 

46: \author{Aram~Kotzinian \\[1.cm] \it

47:  Dipartimento di Fisica Generale, Universit\`a di Torino \\

48:          and \it INFN, Sezione di Torino, Via P. Giuria 1, I-10125 Torino, Italy\\

49: and \it Yerevan Physics Institute, 375036 Yerevan, Armenia \\

50: and \it JINR, 141980 Dubna, Russia\\

51: \it email: aram.kotzinian@cern.ch}

52: 

53: \maketitle

54: 

55: \begin{abstract}

56: 

57: {\tt LEPTO} event generator is modified to describe the azimuthal

58: modulations arising from Cahn and Sivers effects. The comparisons

59: with some existing data in the current fragmentation region of SIDIS

60: are presented. The predictions for Cahn and Sivers asymmetries in

61: the target fragmentation region are presented for SIDIS of 12 GeV

62: electrons off proton target.

63: \end{abstract}

64: 

65: %\newpage

66: \section{Introduction\label{sec:intro}}

67: 

68: In~\cite{akp} the role of parton intrinsic motion in

69: semi-inclusive DIS (SIDIS) processes within QCD parton model has

70: been considered at leading order; intrinsic $\bkt$ is fully taken

71: into account in quark distribution functions and in the elementary

72: processes as well as the hadron transverse momentum, $\bpt$, with

73: respect to fragmenting quark momentum, see

74: Fig.~\ref{fig:planessidis}\footnote{In the following the notations

75: of~\cite{akp} are used.}.

76: %

77: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

78: \begin{figure}[h!]

79: \begin{center}

80: \scalebox{0.4}{\input{planessidis.pstex_t}} \caption{\small Three

81: dimensional kinematics of the SIDIS process.}

82: \label{fig:planessidis}

83: \end{center}

84: \end{figure}

85: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

86: %

87: The average values of $\kt$ for quarks inside protons and $\pt$

88: for final hadrons inside the fragmenting quark jet where fixed by

89: a comparison with data on Cahn effect~\cite{cahn} -- the

90: dependence of the unpolarized cross section on the azimuthal angle

91: between the leptonic and the hadronic planes. The single spin

92: asymmetry (SSA) $A_{UT}^{\sin(\phi_\pi - \phi_S)}$ recently

93: observed by HERMES Collaboration~\cite{hermUT} was successfully

94: described by Sivers mechanism~\cite{siv}. It was also shown that

95: the Sivers distribution functions resulting from this analysis are

96: compatible with the preliminary data from COMPASS

97: collaboration~\cite{compUT}.

98: 

99: The description of SIDIS within standard QCD parton model approach

100: using the distribution and fragmentation functions is valid only

101: in the current fragmentation region, CFR ($x_F>0$) and at high

102: energies. A more general approach allowing to describe SIDIS in

103: the whole kinematic region available for final hadrons is based on

104: the LUND string fragmentation model and is incorporated into {\tt

105: LEPTO} event generator~\cite{lepto}. In the simplest case,

106: corresponding to LO approximation of parton model, event

107: generation in {\tt LEPTO} proceeds in several steps:

108: \begin{enumerate}

109: \item The active quark inside the nucleon is chosen

110: according to the quark density function $f_q(x,Q^2)$,

111: \item The hard scattering kinematics is generated,

112: \item The transverse momentum of the final quark is simulated with Gaussian $\kt$

113: and flat $\varphi$ distributions. Note that the transverse

114: momentum of the final final quark is equal to that of initial

115: quark for leading order hard subprocesses.

116: \item The string fragmentation machinery of {\tt JETSET}

117: program~\cite{JETSET} is applied to form the final hadrons.

118: \end{enumerate}

119: 

120: Within this approach the SIDIS cross section at LO can be

121: expressed as

122: %

123: \be

124:     \label{sidis}

125:     \frac{d^5\sigma^{\ell p \to \ell h X }}{dx \, dQ^2 \, dx_F \,

126:     d^2 \bf P_T} = \sum_q \int {d^2 \bfk _\perp}\; f_{q}(x,k_\perp)

127:     \; \frac{d \hat\sigma ^{\ell q\to \ell q}}{dQ^2} \;

128:     H_{q/N}^h(x,x_F,\bfk_\perp, \bf P_T),

129: \ee

130: %

131: where $\frac{d\hat\sigma ^{\ell q\to \ell q}}{dQ^2}$ is the

132: lepton--quark hard scattering cross section and

133: $H_{q/N}^h(x,x_F,\bfk_\perp, \bf P_T)$ represents the

134: hadronization function of the system formed by struck quark with

135: transverse momentum $\bkt$ and target remnant. In the standard

136: version of {\tt LEPTO} the quark distribution function and the LO

137: lepton--quark cross section are independent of $\varphi$, and,

138: thus the final quarks are uniformly distributed in azimuthal

139: angle. For final hadrons this implies also a uniform azimuthal

140: distribution. However, already in unpolarized SIDIS the observed

141: azimuthal distribution of hadrons is not flat.

142: 

143: In this paper two types of azimuthal modulation at the quark level

144: and their influence on the produced hadron azimuthal distribution

145: will be considered:

146: \begin{itemize}

147: \item{azimuthal modulation of the hard scattering cross section in

148: unpolarized SIDIS (Cahn effect)}

149: \item{azimuthal modulation of the initial quark transverse momentum

150: in SIDIS of unpolarized leptons off the transversely polarized

151: nucleon (Sivers effect)}.

152: \end{itemize}

153: It is possible to incorporate both effects in the {\tt LEPTO}

154: event generator and obtain predictions for azimuthal asymmetries

155: in the whole kinematical region for the final hadrons. The way how

156: the {\tt LEPTO} code is modified to include Cahn and Sivers

157: effects are described in Sec.~\ref{sec:cahn} and

158: Sec.~\ref{sec:sivers}, respectively. In Sec.~\ref{sec:concl} some

159: discussion and conclusions are presented.

160: 

161: 

162: \section{Including Cahn effect in LEPTO\label{sec:cahn}}

163: 

164: 

165: The Cahn effect~\cite{cahn} is a kinematical effect arising due to

166: the presence of nonzero intrinsic transverse momentum of quarks in

167: the nucleon. The underlying physics is very simple. The

168: lepton--quark scattering cross section is given by  the QED

169: expression

170: %

171: \be

172:     d\hat\sigma^{\ell q\to \ell q} \propto \hat s^2+\hat u^2.

173:     \label{lqXsec}

174: \ee

175: %

176: In the general case of non collinear kinematics Mandelstam

177: variables depend on the quark transverse momentum and its

178: azimuthal angle and at order ${\cal O}(\kt/Q)$ one has

179: %

180: \bea

181:     \hat s^2 &=& \frac{Q^4}{y^2} \left( 1 - 4 \frac{\kt}{Q}\,

182:     \sqrt{1-y} \, \cos\varphi \right), \nonumber \\

183:     \hat u^2 &=& \frac{Q^4}{y^2} \, (1-y)^2 \left( 1 - 4 \frac{\kt}{Q}

184:     \, \frac{\cos\varphi}{\sqrt{1-y}} \right). \label{eq:mand}

185: \eea

186: %

187: Then, the lepton--quark elastic scattering cross section is given by

188: %

189: \be

190:     d\hat\sigma^{\ell q\to \ell q} \propto

191:     1-\frac{(2+y)\sqrt{1-y}}{1+(1-y)^2}\frac{\kt}{Q}\cos\varphi.

192:     \label{lqcahn}

193: \ee

194: %

195: Eq. (\ref{lqcahn}) shows that the azimuthal angle of the final

196: quark (and of the string's end associated with the struck quark)

197: is now modulated with amplitude depending on $y, Q$ and $\kt$.

198: 

199: This effect can be introduced in the {\tt LEPTO} event generator

200: at the step 3) of the event generation, when the transverse

201: momentum and azimuthal angle of the scattered quark are generated.

202: To do this the generation of the quark transverse momentum, $\kt$,

203: is left unchanged and then the azimuthal angle is generated

204: according to Eq. (\ref{lqcahn}). This leads to azimuthal

205: modulation of the string axis (axis $\tilde z$ on

206: Fig.~\ref{fig:planessidis}). The momentum conservation means that

207: the transverse momentum of the quark is balanced by that of the

208: target remnant, which in turn means that the azimuthal angle of

209: the target remnant $\varphi_{qq}=\varphi+\pi$. Then, one expects

210: that the azimuthal angle of the hadrons in the target

211: fragmentation region (TFR), $x_F<0$, will be modulated with

212: shifted a phase by $\pi$ with respect to that in CFR.

213: %

214: \begin{figure}[h!]

215: \begin{center}

216: \vspace {-0.5cm}

217:  \includegraphics[width=0.65\linewidth, height=0.55\linewidth]{cahnemc1.eps}

218: \vspace {-0.7cm} \caption{\label{fig:cahnemc} {\small The $x_F$

219: dependence of $\langle \cos \phi_h \rangle / w_1(y)$ for charged

220: hadrons compared with EMC data.}}

221: \end{center}

222: \vspace {-0.5cm}

223: \end{figure}

224: %

225: %

226: \begin{figure}[h!]

227: \begin{center}

228: \vspace {-0.5cm}

229:  \includegraphics[width=0.75\linewidth, height=0.65\linewidth]{cahnjlab1.eps}

230: \vspace {-0.7cm} \caption{\label{fig:cahnjlab1} {\small

231: Predictions of modified {\tt LEPTO} for $x_F$ dependence of

232: $\langle \cos \phi_h \rangle$ for different hadrons produced in 12

233: GeV unpolarized SIDIS process.}}

234: \end{center}

235: \vspace {-0.5cm}

236: \end{figure}

237: %

238: 

239: Data on azimuthal dependencies of SIDIS covering a large $x_F$

240: range have been obtained by the EMC Collaboration~\cite{emc} for a

241: beam energy of 280 GeV. The $x_F$ dependence of $\langle \cos

242: \phi_h \rangle /w_1(y),$ where

243: $w_1(y)=(2-y)\sqrt{1-y}/\left(1+(1-y)^2\right)$, obtained by using

244: modified {\tt LEPTO} for EMC kinematics are presented in

245: Fig.~\ref{fig:cahnemc} together with data points from~\cite{emc}.

246: The simulations has been done with LO setting of {\tt LEPTO}

247: (LST(8)=0) and with values of the parameters describing intrinsic

248: $k_T$ (PARL(3)=0.5) and fragmentation $p_T$ (PARL(21)=0.45) as

249: adopted in~\cite{akp}.

250: 

251: The predictions of modified {\tt LEPTO} for $\langle \cos \phi_h

252: \rangle$ of different hadron ($\pi^+$, $\pi^-$, $\pi^0$ and $p$)

253: produced in SIDIS on a proton target at future CEBAF 12 GeV

254: facility at JLab~\cite{jlab12} are presented in

255: Fig.~\ref{fig:cahnjlab1}. One can see from Fig.~\ref{fig:cahnemc}

256: and Fig.~\ref{fig:cahnjlab1} that the predicted mean value of

257: $\cos \phi_h$ in the CFR is negative $\langle \cos \phi_h

258: \rangle_{CFR}<0$, while in the TFR is positive $\langle \cos

259: \phi_h \rangle_{TFR}>0$, as suggested by arguments based on

260: transverse momentum conservation.

261: 

262: 

263: \section{Including Sivers effect in LEPTO\label{sec:sivers}}

264: 

265: 

266: The azimuthal modulation of the string transverse momentum in the

267: previous section was due to Cahn effect -- the dependence of the

268: non planar lepton-quark scattering cross section on the quark

269: azimuth. The quark distribution, $f_q(x,k_\perp)$ itself is

270: independent of quark azimuthal angle.

271: 

272: The situation is different when one considers SIDIS on a

273: transversely polarized nucleon. Now a correlation between

274: transverse momentum of quark and target transverse polarization is

275: possible -- the so called Sivers effect~\cite{siv}. For quite some

276: time it was believed that this correlation is forbidden because of

277: T-invariance of the strong interactions. However the spectator

278: model calculations~\cite{bhs} demonstrated that there exists a

279: nonzero SSA in SIDIS when the final state interaction between

280: struck quark and target remnant is taken into account. Then, the

281: effective description of this SSA is possible within QCD

282: factorized approach by introducing a new distribution function --

283: the Sivers function (for further discussion see, for

284: example,~\cite{metz}).

285: 

286: The unpolarized quark (and gluon) distributions inside a

287: transversely polarized proton (generically denoted by $\pup$, with

288: $\pdown$ denoting the opposite polarization state) can be written

289: as:

290: %

291: %

292: \be

293:     f_ {q/\pup} (x,\bfk_\perp) = f_ {q/p} (x,\kt) + \frac{1}{2} \,

294:     \Delta^N f_ {q/\pup}(x,\kt)  \; {\bfS} \cdot (\hat {\bfP} \times

295:     \hat{\bfk}_\perp)\; , \label{poldf}

296: \ee

297: %

298: where $\bfP$ and $\bfS$ are respectively the proton momentum and

299: transverse polarization vector, and $\bfk_\perp$ is the parton

300: transverse momentum; transverse refers to the proton direction.

301: Eq. (\ref{poldf}) implies

302: %

303: \bea

304:     \nonumber &&f_ {q/\pup} (x,\bfk_\perp) + f_ {q/\pdown}

305:     (x,\bfk_\perp) =

306:     2 f_ {q/p} (x,\kt)\;, \\

307:     &&f_ {q/\pup} (x,\bfk_\perp) - f_ {q/\pdown} (x,\bfk_\perp) =

308:     \Delta^N f_ {q/\pup}(x,\kt)\;{\bfS} \cdot (\hat{\bfP}  \times \hat

309:     {\bfk}_\perp)\;, \label{sivf}

310: \eea

311: %

312: where $f_ {q/p} (x,\kt)$ is the unpolarized parton density and

313: $\Delta^N f_ {q/\pup}(x,\kt)$ is referred to as the Sivers function.

314: Notice that, as requested by parity invariance, the scalar quantity

315: $\bfS \cdot (\hat{\bfP} \times \hat {\bfk}_\perp)$ singles out the

316: polarization component perpendicular to the $\bfP-\bfk_\perp$ plane.

317: For a proton moving along $-z$ and a generic transverse polarization

318: vector $\bfS = |\bfS|\,(\cos\phi_S, \sin\phi_S, 0)$ (see Fig.

319: \ref{fig:planessidis}) one has:

320: %

321: \be

322:     \bfS \cdot (\hat{\bfP}  \times \hat{\bfk}_\perp) = |\bfS| \,

323:     \sin(\varphi-\phi_S) \equiv |\bfS| \, \sin\phi_{Siv} \,,

324: \ee

325: %

326: where $(\varphi-\phi_S) = \phi_{Siv}$ is the Sivers

327: angle.

328: 

329: In~\cite{akp} the Sivers function for each light quark flavor $q=u,d$

330: are parameterized in the following factorized form:

331: %

332: \be

333:     \Delta^N f_ {q/\pup}(x,\kt) = 2 \, {\cal N}_q(x) \, h(\kt) \,

334:     f_ {q/p} (x,\kt)\; , \label{sivfac}

335: \ee

336: %

337: where

338: %

339: \bea

340:     &&{\cal N}_q(x) =  N_q \, x^{a_q}(1-x)^{b_q} \,

341:     \frac{(a_q+b_q)^{(a_q+b_q)}}{a_q^{a_q} b_q^{b_q}}\; ,

342:     \label{siversx} \\

343:     &&h(\kt) = \sqrt{2e} \, \frac{\kt} {M} \, e^{-\kt^2/M^{2}}\;  ,

344:     \label{siverskt}

345: \eea

346: %

347: where $N_q$, $a_q$, $b_q$ and $M$ (GeV/$c$) are parameters. Then

348: Eq.~(\ref{poldf})can be rewritten as

349: %

350: \be

351:     f_ {q/\pup} (x,\bfk_\perp) = f_ {q/p} [x,\kt)(1+|\bfS|{\cal N}_q(x)h(\kt) \,

352:     \sin\phi_{Siv}].

353:     \label{sivmod}

354: \ee

355: %

356: 

357: Again, the Sivers effect is incorporated into {\tt LEPTO} at the

358: stage 3) of the event generation in the same way as for the Cahn

359: effect but now the azimuthal angle is generated according to

360: Eq.~(\ref{sivmod}). For simulations the following set of

361: parameters compatible with those obtained in~\cite{akp} have been

362: used: $N_u=N_{\bar u}=0.5$, $N_d=N_{\bar d}=-0.2$, $a_q=0.3$,

363: $b_q=2$ and $M^2=0.36$ (GeV/c)$^2$.

364: 

365: In Fig.~\ref{fig:sivherm} the results of simulation for HERMES

366: experimental conditions are compared with observed Sivers

367: asymmetries~\cite{hermUT}.

368: 

369: \begin{figure}[h!]

370: \begin{center}

371: \vspace {-0.5cm}

372:  \includegraphics[width=0.75\linewidth, height=0.65\linewidth]{sivherm.eps}

373: \vspace {-0.7cm} \caption{\label{fig:sivherm} {\small HERMES data on

374: $A_{UT}^{\sin(\phi_\pi-\phi_S)}$ \cite{hermUT} for scattering off a

375: transversely polarized proton target. The curves are the results of

376: simulations obtained with modified {\tt LEPTO}}.}

377: \end{center}

378: \vspace {-0.5cm}

379: \end{figure}

380: 

381: Future facilities as Electron Ion Colliders or upgraded JLab will

382: have larger kinematic coverage and will offer the possibility of

383: studying the Sivers effect also with hadrons produced in the TFR.

384: As an example, the simulations have been done for 12 GeV electron

385: SIDIS of a proton target. The DIS cut $Q^2>1 (Gev/c)^2$ and $W^2>4

386: Gev^2$ and a cut on the produced hadron transverse momentum

387: $P_T>0.05$ GeV/c was imposed. The predictions for $x_F$, $x$ and

388: $P_T$ dependencies for JLab kinematics are presented on the

389: Fig.~\ref{fig:sivxf}, Fig.~\ref{fig:sivxb} and

390: Fig.~\ref{fig:sivpt}, respsctively.

391: 

392: \begin{figure}[h!]

393: \begin{center}

394: \vspace {-0.5cm}

395:  \includegraphics[width=0.75\linewidth, height=0.65\linewidth]{sivjlabl_xf.eps}

396: \vspace {-0.7cm} \caption{\label{fig:sivxf} {\small Predicted

397: dependence

398:     of $A_{UT}^{\sin(\varphi_h-\varphi_S)}$ on $x_F$ for different

399:     hadrons produced in SIDIS of 12 GeV electrons off a transversely polarized

400:     proton target.}}

401: \end{center}

402: \vspace {-0.5cm}

403: \end{figure}

404: 

405: \begin{figure}[h!]

406: \begin{center}

407: \vspace {-0.5cm}

408:  \includegraphics[width=0.75\linewidth, height=0.65\linewidth]{sivjlabl_xb.eps}

409: \vspace {-0.7cm} \caption{\label{fig:sivxb} {\small Predicted

410: dependence

411:     of $A_{UT}^{\sin(\varphi_h-\varphi_S)}$ on $x$ for different

412:     hadrons produced in the TFR ($x_F<-0.1$) of SIDIS of 12 GeV electrons off

413:     a transversely polarized proton target.}}

414: \end{center}

415: \vspace {-0.5cm}

416: \end{figure}

417: 

418: 

419: \begin{figure}[h!]

420: \begin{center}

421: \vspace {-0.5cm}

422:  \includegraphics[width=0.75\linewidth, height=0.65\linewidth]{sivjlabl_pt.eps}

423: \vspace {-0.7cm} \caption{\label{fig:sivpt} {\small Predicted

424: dependence

425:     of $A_{UT}^{\sin(\varphi_h-\varphi_S)}$ on $p_T$ for different

426:     hadrons produced in the TFR ($x_F<-0.1$) of SIDIS of 12 GeV electrons off

427:     a transversely polarized proton target.}}

428: \end{center}

429: \vspace {-0.5cm}

430: \end{figure}

431: 

432: \section{Discussion and Conclusions\label{sec:concl}}

433: 

434: In this article the way of modifying the standard {\tt LEPTO}

435: event generator in order to include the azimuthal asymmetries

436: arising from Cahn and Sivers effects is described. Only LO effects

437: have been taken into account. The azimuthal modulations for Cahn

438: and Sivers effects have different origins. In the case of Cahn

439: effect the initial quark transverse momentum is independent of

440: azimuthal angle but the hard scattering cross section in a non

441: planar kinematics depends on the final quark azimuthal angle. In

442: the case of Sivers effect already the initial quark transverse

443: momentum has an azimuthal modulation. The azimuthal asymmetries

444: are introduced in both cases by changing the struck quark/string

445: azimuthal distribution during event generation. The hadronization

446: part of program ({\tt JETSET}) is left unchanged. The possible

447: influence of the higher twist distribution functions as well as

448: possible modifications of hadronization in the case of polarized

449: target~\cite{ak} have been ignored.

450: 

451: The advantage of this MC based approach compared to standard QCD

452: factorized approach is in the full coverage of produced hadron

453: phase space. The modified generator will be useful for complete MC

454: simulations of experiments including Cahn and Sivers effects both

455: in the CFR and in the TFR and also for global analysis of these

456: effects.

457: 

458: Figs.~\ref{fig:cahnemc} and~\ref{fig:sivherm} demonstrate that the

459: modified {\tt LEPTO} event generator well describing the data in

460: the CFR both for Cahn and Sivers asymmetries. The description of

461: Cahn effect in the TFR looks unsatisfactory. This discrepancy can

462: be explained either by some unaccounted contributions in the TFR

463: or by insufficient precision of experimental data. One can notice

464: from the experimental points in Fig.~\ref{fig:cahnemc} that the

465: integrated value of $\langle \cos \phi_h \rangle$ for charged

466: hadrons in the CFR is not compensated by that in TFR. It seems

467: improbable that this imbalance can be compensated by larger values

468: of $\langle \cos \phi_h \rangle$ of neutral hadrons at $x_F \simeq

469: -1$.

470: 

471: The new high statistic measurements will allow to check the

472: predictions of the approach presented here and better understand

473: the effects of the quark intrinsic transverse momentum and

474: hadronization mechanism in SIDIS.

475: 

476: 

477: \section*{Acknowledgements}

478: 

479: 

480: The author express his gratitude to M.~Anselmino, A.~Prokudin for

481: discussions and also to NUCLEOFIT group members of the General

482: Physics Department `A.~Avogadro' of the Turin University for

483: interest in this work.

484: 

485: \begin{thebibliography}{99}

486: 

487: \bibitem{akp} M.~Anselmino {\it et al.}, arXiv:hep-ph/0501196, to be published in \PR

488: 

489: \bibitem{cahn}

490:   R.N. Cahn, \PL{B78} (1978) 269; \PR{D40} (1989) 3107.

491: 

492: \bibitem{hermUT}

493:   HERMES Collaboration, A. Airapetian {\it et al.}, \PRL{94} (2005) 012002,

494:   arXive:hep-ex/0408013.

495: 

496: \bibitem{siv}

497:   D. Sivers, \PR{D41} (1990) 83; {\bf D43} (1991) 261.

498: 

499: \bibitem{compUT}

500:   COMPASS Collaboration, V.~Y.~Alexakhin {\it et al.},

501:   %``First measurement of the transverse spin asymmetries of the deuteron in

502:   %semi-inclusive deep inelastic scattering,''

503:   arXiv:hep-ex/0503002, to be published in \PRL

504: 

505: \bibitem{lepto}

506:   G.~Ingelman, A.~Edin and J.~Rathsman,

507:   {\it Comp.\ Phys.\ Commun.} {\bfseries 101} 108 (1997).

508: 

509: \bibitem{JETSET}

510:   T.~Sj\"ostrand, {\it Comp.\ Phys.\ Commun.} {\bfseries 39} 347 (1986),

511: {\bfseries 43} 367 (1987);\\

512: T.~Sj\"ostrand, {\it PYTHIA 5.7 and JETSET 7.4: Physics

513:   and Manual}, arXiv:hep-ph/9508391; \\

514:  T.~Sj\"ostrand, {\it PYTHIA 6.2: Physics and Manual},

515:  arXiv:hep-ph/0108264.

516: 

517: \bibitem{emc}

518:   EMC Collaboration, M.~Arneodo {\it et al.}, {\it Z.\ Phys.}

519:   {\bf C34} (1987) 277.

520: 

521: \bibitem{jlab12}

522: Pre-Conceptual Design Report,

523: http://www.jlab.org/12GeV/collaboration.html.

524: 

525: \bibitem{bhs}

526: S.~J.~Brodsky, D.~S.~Hwang and I.~Schmidt, \PL{B530} (2002) 99.

527: 

528: \bibitem{metz}

529:   A.~Metz,

530:   %``Present understanding of the nucleon spin structure,''

531:   arXiv:hep-ph/0412156.

532: 

533: \bibitem{ak}

534: A.~Kotzinian, arXiv:hep-ph/0410093.

535: 

536: \end{thebibliography}

537: 

538: \end{document}

539: