hep-ph0302158/ys1.tex
1: 
2: %Revtex4
3: \documentclass[preprint,showpacs,prd,amsmath,amssymb]{revtex4}
4: %\documentclass[twocolumn,superscriptaddress,showpacs,prd,amsmath,amssymb]{revtex4}
5: \usepackage{graphicx}
6: %\usepackage{dcolumn}
7: 
8: \renewcommand{\thefootnote}{\fnsymbol{footnote}}
9: \renewcommand{\bar}[1]{\overline{#1}}
10: \newcommand{\ie}{{\it i.e.}}
11: \newcommand{\btt}[1]{{\tt$\backslash$#1}}
12: \renewcommand{\d}{{\mathrm d}}
13: 
14: %\textwidth 6.0in \textheight 8.6in
15: \renewcommand{\baselinestretch}{1.5}
16: 
17: \begin{document}
18: 
19: 
20: \preprint{USM-TH-135}
21: 
22: \title{Mass Suppression in Octet Baryon Production}
23: 
24: 
25: \author{Jian-Jun Yang}
26: \email{Jian-Jun.Yang@physik.uni-regensburg.de}
27: \affiliation{Department of Physics, Nanjing Normal
28: University, Nanjing 210097, China}
29: \affiliation{Institut f\"ur Theoretische Physik,
30: Universit\"at Regensburg, D-93040 Regensburg, Germany}
31: 
32: 
33: 
34: \author{Ivan Schmidt}
35: \email{Ivan.Schmidt@fis.utfsm.cl}
36: \affiliation{Departamento de F\'\i sica, Universidad
37: T\'ecnica Federico Santa Mar\'\i a, Casilla 110-V, Valpara\'\i so, Chile}
38: 
39: 
40: \begin{abstract}
41: 
42: There is a striking suppression of the cross section for
43: production of octet baryons in $e^+ e ^-$ annihilation, as the
44: mass of the produced hadron increases. We present a simple
45: parametrization for the fragmentation functions into octet baryons
46: guided by two input models: the SU(3) flavor symmetry part is
47: given by a quark-diquark model, and the baryon mass suppression
48: part is inspired by the string model. We need only {\it eight} free
49: parameters to describe the fragmentation functions for all octet
50: baryons. These free parameters are determined by a fit to the
51: experimental data of octet baryon production in $e^+ e ^-$
52: annihilation. Then we apply the obtained fragmentation functions
53: to predict the cross section of the octet baryon production in
54: charged lepton DIS and find consistency with the available
55: experimental data. Furthermore, baryon production in $pp$
56: collisions is suggested to be an ideal domain to check the
57: predicted mass suppression.
58: 
59: \end{abstract}
60: 
61: \pacs{14.20.Jn, 13.60.Rj, 13.65.+i, 13.87.Fh}
62: 
63: \maketitle
64: 
65: 
66: 
67: 
68: \section{Introduction}
69: 
70: It is well known that there is an SU(3) flavor symmetry between
71: the wave functions of the octet baryons. Very recent lattice QCD
72: calculations also suggest that the $\Lambda$ and proton quark
73: structures can be related by an SU(3)
74: transformation~\cite{Lattice}. Unfortunately it is not possible to
75: investigate the parton distributions of the octet hyperons by
76: means of structure functions in deep inelastic scattering (DIS),
77: since they cannot be used as a target due to their short
78: life-time. Also one obviously cannot produce a beam of
79: charge-neutral hyperons such as $\Lambda$. Therefore there is no
80: experimental information on the relation between the parton
81: distributions of octet hyperons and those of the nucleon. However,
82: it has been proposed that the quark fragmentation functions might
83: be related to the corresponding quark distributions by a so called
84: Gribov-Lipatov relation~\cite{GLR,Bro97}. Therefore we can explore
85: the relation between quark distributions of octet baryons by means
86: of hyperon production from quark fragmentation. Actually great
87: progress has been made along this direction
88: recently~\cite{MSSY,MSY}.
89: 
90: A specific regularity in baryon production rates in $e^+e^-$
91: annihilation has been recently noticed by Chliapnikov and
92: Uvarov~\cite{CU95}. However, to our knowledge, there has not been
93: a detailed investigation on the relation among the octet baryon
94: fragmentation functions. The data shows a striking suppression of
95: the cross section as the mass of the produced hadron increases,
96: which is puzzling since naive SU(3) would suggest that they should
97: be comparable. For example, once an s-quark is produced then one
98: could think that the $\Sigma :\Lambda$ ratio would be $3:1$, since
99: the $qq$ pair is statistically three times more likely to be in an
100: isospin $I=1$ state than in isospin $I=0$. Nevertheless the
101: experimental data shows just the opposite trend, with a lower
102: cross section for $\Sigma$ production (see Fig. \ref{ys1f1}).
103: Apparently something in the fragmentation process makes it much
104: easier for the s-quark to pick up an isosinglet $qq$ than an
105: isotriplet.
106: 
107: In this paper we will not provide a detailed physical explanation
108: of the suppression. Our, more modest, purpose is to give a
109: consistent simple parameterization of all the fragmentation
110: functions into octet baryons, with a rather small number of free
111: parameters. Nevertheless, since the parameterizations that we
112: obtain are inspired by well known input models, they can be
113: considered a first step into a physical understanding of these
114: processes. For every octet baryon, there are 18 (unpolarized,
115: longitudinally polarized and transversely polarized)
116: quark/antiquark fragmentation functions. As a sensible
117: parametrization of a fragmentation function usually needs at least
118: 3 parameters, a lot of experimental data are needed in order to
119: fix all of them. We plan to constrain the shape of fragmentation
120: functions with the help of some models in order to reduce the
121: number of free parameters. Models such as those using strings and
122: shower algorithms~\cite{Barger} still involve many parameters, and
123: therefore with these models it is difficult to obtain a clear
124: relation among the fragmentation functions for various octet
125: baryons. On the other hand, the diquark model given in
126: Ref.~\cite{Nza95} can provide us with SU(3) flavor symmetry
127: relations for octet baryon fragmentation functions. An
128: investigation of the spin structure of the diquark fragmentation
129: functions can be found in previous
130: publications~\cite{Nza95,a01,a06}. It was found that the spin
131: structure of the diquark model fragmentation functions for the
132: $\Lambda$ is supported by the available experimental
133: data~\cite{a01,a06}, which indicates that the diquark model works
134: well in describing the fragmentation functions. In the present
135: work, we will retain the flavor and spin structure of
136: fragmentation functions as given by the diquark  model and
137: concentrate on the relation between fragmentation functions of
138: various octet baryons. We will emphasize the mass suppression
139: effect in hyperon production.
140: 
141: 
142: The paper is organized as follows. In Sec.~II, we briefly describe
143: our ansatz of the fragmentation functions for all octet baryons
144: based on the diquark model and the string model. Using only eight
145: free parameters, we relate the fragmentation functions for all
146: octet baryons to each other. The free parameters of the model are
147: determined based on the experimental data on octet baryons
148: production in $e^+e^-$ annihilation, where the mass suppression
149: effect in hyperon production is very important in order to
150: understand the available experimental data. In Sec.~III, we
151: propose a possible cross check of the mass suppression via  octet
152: baryon production in charged lepton DIS. In Sec.~IV, we give
153: predictions for cross sections of baryon production in $pp$
154: collisions with the obtained fragmentation functions for the octet
155: baryons. Finally, we give a summary with our conclusions in
156: Sec.~V.
157: 
158: \section{Quark fragmentation functions for the octet baryons}
159: 
160: Recently, it was found that a simple diquark model can be used to
161: describe quite accurately the octet baryon fragmentation
162: functions~\cite{Nza95,a01,a06}. The parameters of the model were
163: determined by fitting the experimental data of octet baryon
164: production in $e^+e^-$ annihilation. In this work, we adopt an
165: alternative approach and present a parametrization based on the
166: spin and flavor structure predicted by the diquark model. We focus
167: our attention on the relation between the fragmentation functions
168: of various octet baryons.
169: 
170: In contrast to the nucleon parton distributions which are well
171: determined by  experimental data, we have much less information on
172: fragmentation functions for the octet baryons. For this reason, we
173: constrain the fragmentation functions with the help of some models
174: in order to reduce the number of free parameters. In particular
175: the diquark model~\cite{Nza95} has a clear physical motivation and
176: needs only a few parameters.
177: 
178: Within the framework of the diquark model~\cite{Nza95}, the
179: unpolarized valence quark to  proton fragmentation functions can
180: be expressed as
181: 
182: \begin{equation}
183: D_{u_v}^p(z)= \frac{1}{2} a_S^{(u/p)}(z)
184: + \frac{1}{6} a_V^{(u/p)}(z),
185: \end{equation}
186: 
187: \begin{equation}
188: D_{d_v}^p(z)= \frac{1}{3} a_V^{(d/p)}(z),
189: \end{equation}
190: where $a_D^{(q/p)}(z)$  ($D=S$ or $V$) is the probability of
191: finding a quark $q$ splitting into the proton $p$ with
192: longitudinal momentum fraction $z$ and emitting a scalar ($S$) or
193: axial vector ($V$) anti-diquark. The form factors for scalar and
194: axial vector diquark are customarily taken to have the same form
195: 
196: \begin{equation}
197: \phi (k^2)=N \frac{k^2-m^2_q}{(k^2- \Lambda_0^2)^2},\label{form}
198: \end{equation}
199: with a normalization constant $N$ and a mass parameter
200: $\Lambda_0$. The value $\Lambda_0=500~\rm{MeV}$ is usually adopted
201: in numerical calculations. In (\ref{form}), $m_q$ and $k$ are the
202: mass and  the momentum of the fragmenting quark $q$, respectively.
203: According to Ref.~\cite{Nza95}, in the quark-diquark model
204: $a_D^{(q/p)}(z)$ can be expressed as
205: 
206: \begin{equation}
207: a_D^{(q/p)}(z)=\frac{N^2 z^2 (1-z)^3}{64 \pi^2}
208: \frac {[2 ( M_p + m_q z)^2+
209:  R^2(z)]}{R^6(z)}\label{ad}
210: \end{equation}
211: with
212: 
213: \begin{equation}
214: R(z)=\sqrt{z m_D^2-z(1-z)\Lambda_0^2+(1-z)M_p^2},
215: \end{equation}
216: where $M_p$ and $m_D$~($D=S$ or $V$) are the mass of the proton
217: and diquark, respectively. We choose the values for the diquark
218: masses to be $m_S=900~ \rm{MeV}$ and $m_V=1100~ \rm{MeV}$, for
219: scalar and axial vector diquark states, respectively. The quark
220: masses are taken as $m_u=m_d= 350~ \rm{MeV}$.
221: 
222: Similarly, the longitudinally and transversely  polarized quark to
223: proton  fragmentation functions can be written  as
224: 
225: \begin{equation}
226: \Delta D_{u_v}^p(z)
227: = \frac{1}{2} \tilde{a}_S^{(u/p)}(z)
228: - \frac{1}{18} \tilde{a}_V^{(u/p)}(z),
229: \end{equation}
230: 
231: \begin{equation}
232: \Delta D_{d_v}^p(z)= -\frac{1}{9} \tilde{a}_V^{(d/p)}(z),
233: \label{Dsv}
234: \end{equation}
235: 
236: \begin{equation}
237: \delta D_{u_v}^p(z)
238: = \frac{1}{2} \hat{a}_S^{(u/p)}(z)
239: - \frac{1}{18} \hat{a}_V^{(u/p)}(z),
240: \end{equation}
241: 
242: \begin{equation}
243: \delta D_{d_v}^p(z)= -\frac{1}{9} \hat{a}_V^{(d/p)}(z),
244: \label{dsv}
245: \end{equation}
246: with
247: 
248: \begin{equation}
249: \tilde{a}_D^{(q/p)}(z)=\frac{N^2 z^2 (1-z)^3}{64 \pi^2}
250: \frac {[2( M_p + m_q z)^2-R^2(z)]}{R^6(z)},\label{adb}
251: \end{equation}
252: and
253: 
254: \begin{equation}
255: \hat{a}_D^{(q/p)}(z)=\frac{N^2 z^2 (1-z)^3}{32 \pi^2}
256: \frac {( M_p + m_q z)^2}{R^6(z)},\label{adt}
257: \end{equation}
258: for $D=S$ or $V$. Here we are not interested in the absolute
259: magnitude of the fragmentation functions but in the flavor and
260: spin structure of them, which is given by the diquark model. In
261: order to extract the flavor and spin structure information, we
262: introduce flavor structure ratios
263: 
264: \begin{equation}
265: F_V^{(u/d)}(z)=\frac{a_V^{(u/p)}(z)}{a_V^{(d/p)}(z)},
266: \end{equation}
267: 
268: \begin{equation}
269: F_M^{(u/d)}(z)=\frac{a_S^{(u/p)}(z)}{a_V^{(d/p)}(z)},
270: \end{equation}
271: and spin structure ratios
272: 
273: \begin{equation}
274: W_D^{(q/p)}(z)=\frac{\tilde{a}_D^{(q/p)}(z)}{a_D^{(q/p)}(z)},
275: \end{equation}
276: 
277: \begin{equation}
278: \hat{W}_D^{(q/p)}(z)=\frac{\hat{a}_D^{(q/p)}(z)}{a_D^{(q/p)}(z)},
279: \end{equation}
280: with $D=S$ or  $V$.
281: Then we can use the fragmentation function $D_{d_v}^p (z)$ to
282: express all other unpolarized and polarized fragmentation functions
283: for the proton as follows
284: 
285: 
286: \begin{equation}
287: D_{u_v}^p (z)=\frac{1}{2}[ F_V^{(u/d)}(z) + 3 F_M^{(u/d)}(z) ]
288: D_{d_v}^p (z), \label{eq16}
289: \end{equation}
290: 
291: \begin{equation}
292: \Delta D_{u_v}^p (z)=\frac 32 [ W_S^{(u/p)}(z)  F_M^{(u/d)}(z) -
293: \frac{1}{9} W_V^{(u/p)}(z)  F_V^{(u/d)}(z) ]
294: D_{d_v}^p (z),
295: \end{equation}
296: 
297: \begin{equation}
298: \Delta D_{d_v}^p (z)= -\frac{1}{3} W_V^{(d/p)}(z) D_{d_v}^p (z),
299: \end{equation}
300: 
301: \begin{equation}
302: \delta D_{u_v}^p (z)=\frac 32 [ \hat{W}_S^{(u/p)}(z)  F_M^{(u/d)}(z) -
303: \frac{1}{9} \hat{W}_V^{(u/p)}(z)  F_V^{(u/d)}(z) ]
304: D_{d_v}^p (z),
305: \end{equation}
306: and
307: 
308: \begin{equation}
309: \delta D_{d_v}^p (z)= -\frac{1}{3} \hat{W}_V^{(d/p)}(z) D_{d_v}^p (z).
310: \label{eq20}
311: \end{equation}
312: 
313: The spin structure of the quark-diquark fragmentation functions
314: for the $\Lambda$ has been studied before~\cite{a01,a06}, and it
315: is supported by the available experimental data on $\Lambda$
316: production in various processes~\cite{ALEPH96, DELPHI95, OPAL97,
317: HERMES, E665, NOMAD}. In this work, we retain the flavor and spin
318: structure of the fragmentation functions suggested by the diquark
319: model.
320: 
321: From the above analysis, we find out that the essential ingredient
322: is to choose a suitable shape for the function $D^p_{d_v}(z)$ from
323: which the other valence fragmentation functions can then be
324: deduced. We could use the expression of $D^p_{d_v}(z)$ coming from
325: the diquark approach and  fix the parameters of the model by a fit
326: to the experimental data as it was done in Ref.~\cite{a06}.
327: Another way is to use a commonly accepted parametrization form
328: such as
329: 
330: \begin{equation}
331: D_{d_v}^p (z, Q^2_0) = N_v z ^{\alpha_v} (1-z) ^{\beta_v}, \label{fit1}
332: \end{equation}
333: with the exponents $\alpha_v$ and $\beta_v$ at an initial scale
334: $Q^2_0$. Previous work~\cite{a01,a06} indicates  that compatible
335: results can be obtained in both ways. In our present analysis, we
336: adopt the later approach, with the analytical expression
337: (\ref{fit1}) for $D^p_{d_v}(z)$, since this simple parametrization
338: can be easily used later for other purposes. In addition, the
339: diquark model fragmentation functions are easier to describe in
340: the large $z$ region where the valence quark contribution
341: dominates. In the small $z$ region, the sea contribution is
342: difficult to include in the framework of the diquark model.
343: Nevertheless, we also adopt a similar functional form
344: 
345: 
346: \begin{equation}
347: D_{q_s}^p (z, Q^2_0) = D_{\bar{q}}^p (z, Q^2_0)= N_s z ^{\alpha_s}
348: (1-z) ^{\beta_s} \label{fit2}
349: \end{equation}
350: to parameterize fragmentation functions of the  sea quark
351: $D_{q_s}^p (z)$ and antiquark $D_{\bar{q}}^p (z)$ for $q=u, d, s$
352: at the initial scale $Q^2_0$. For simplicity, we take the same
353: initial parametrization for the spin independent gluon and the sea
354: quark fragmentation functions, and moreover we assume that $\Delta
355: D_g^p$, $\delta D_g^p$, $\Delta D_{q_s(\bar{q})}^p$, and $\delta
356: D_{q_s(\bar{q})}^p$ at the initial scale are zero and that they
357: are only generated by QCD evolution.
358: 
359: Hence, the input unpolarized and polarized quark to proton
360: fragmentation functions can be written as
361: 
362: \begin{equation}
363: D_{q}^{p[\rm{SU}(3)]} (z, Q^2_0) = D_{q_v}^p (z, Q^2_0) + D_{q_s}^p (z, Q^2_0),
364: \end{equation}
365: 
366: \begin{equation}
367: \Delta D_{q}^{p[\rm{SU}(3)]} (z, Q^2_0) = \Delta D_{q_v}^p (z, Q^2_0),
368: \end{equation}
369: and
370: 
371: \begin{equation}
372: \delta D_{q}^{p[\rm{SU}(3)]} (z, Q^2_0) = \delta D_{q_v}^p (z, Q^2_0).
373: \end{equation}
374: 
375: Now we deduce the fragmentation functions
376: $D_{q(g)}^{B[\rm{SU}(3)]}$ for all other octet baryons $B$ by
377: SU(3) symmetry  at the initial scale $Q^2_0$. More specifically,
378: we have
379: 
380: 
381: \begin{equation}
382: \begin{array}{lllc}
383: D_u^{p[\rm{SU}(3)]}=D_d^{n[\rm{SU}(3)]}=
384: D_u^{{\Sigma^+}[\rm{SU}(3)]}=D_d^{{\Sigma^-}[\rm{SU}(3)]}
385: =D_s^{{\Xi^-}[\rm{SU}(3)]}=D_s^{{\Xi^0}[\rm{SU}(3)]}\\
386: \mbox{}\hspace{1.5cm}=\frac{2}{3}D_u^{{\Lambda}[\rm{SU}(3)]}+\frac{4}{3}
387: D_s^{{\Lambda}[\rm{SU}(3)]} =2 D_u^{{\Sigma^0}[\rm{SU}(3)]}=2 D_d
388: ^{{\Sigma^0}[\rm{SU}(3)]};\\
389: D_d^{p[\rm{SU}(3)]}=D_u^{n[\rm{SU}(3)]}=D_s^{{\Sigma^+}[\rm{SU}(3)]}=
390: D_s^{{\Sigma^-}[\rm{SU}(3)]}=D_d^{{\Xi^-}[\rm{SU}(3)]}
391: =D_u^{{\Xi^0}[\rm{SU}(3)]}\\
392: \mbox{}\hspace{1.5cm}=\frac{4}{3}D_u^{{\Lambda}[\rm{SU}(3)]}-
393: \frac{1}{3}D_s^{{\Lambda}[\rm{SU}(3)]} =D_s
394: ^{{\Sigma^0}[\rm{SU}(3)]},
395: \end{array}
396: \label{su3}
397: \end{equation}
398: with similar relations for the polarized fragmentation functions.
399: We assume that the sea quark fragmentation functions also have the
400: above SU(3) relations. In principle the diquark model can also be
401: used to partly reflect the SU(3) flavor symmetry breaking effect
402: if the differences in the quark, anti-diquark, and baryon masses
403: are taken into account in the probabilities
404: $a_D^{(q/B)}(z)$~\cite{a06} for a quark $q$ fragmenting into the
405: baryon $B$. However the SU(3) symmetry breaking effect due to this
406: difference in quark, diquark and baryon masses in the diquark
407: model is too weak to explain the experimentally measured values
408: for the average hadronic multiplicities per hadronic $e^+e^-$
409: annihilation event~\cite{PDG00}, where hyperon production is
410: significantly suppressed as compared with proton production.
411: Actually, we find that the cross sections for $\Lambda$, $\Sigma$,
412: and $\Xi$ baryons in $e^+e^-$ annihilation would be overestimated
413: by up to two orders of magnitude if we only considered this SU(3)
414: symmetry breaking in the framework of the diquark model. We have
415: to search for another possible source of the suppression effect in
416: hyperon production. In Ref.~\cite{Rastogi}, a description of the
417: strangeness suppression effect was proposed by putting a
418: suppression factor in the $u$, $d$, and sea quark fragmentation
419: functions for baryons containing a valence $s$ quark (and a
420: further overall suppression factor for baryons containing two $s$
421: quarks). In our present analysis, we will consider an alternative
422: suppression mechanism due to the hyperon masses which is inspired
423: by the string model~\cite{String}. For simplicity, we do not
424: include  SU(3) symmetry breaking caused by the difference in
425: quark, diquark and baryon masses in the diquark model itself since
426: this is small. We introduce an additional mass suppression factor
427: for the SU(3) symmetric fragmentation functions of the  diquark
428: model. This overall mass suppression factor should not alter
429: significantly the flavor  and spin structure of the fragmentation
430: functions as given by the diquark model. More specifically, we
431: assume that the quark $q$ to baryon  $B$ fragmentation function
432: can be expressed as follows:
433: 
434: \begin{equation}
435: D_{q}^B(z) = D_{q}^{B[\rm{SU}(3)]}(z) (2J+1) \left\{ 1 +
436: \frac{|S|}{(2I+1)}\right \} \exp [-b M_B^2/z^c] \label{ssp}
437: \end{equation}
438: where $S$, $I$, $J$ and $M_B$ are the strangeness, isospin,
439: spin and mass of the octet baryon $B$. The  term within the
440: curly brackets is a strangeness modification factor. The mass
441: suppression factor is inspired by the string model~\cite{String},
442: and $D_{q}^{B[SU(3)]}(z)$ is the SU(3) diquark model fragmentation
443: functions for the octet baryon $B$.
444: 
445: To summarize, our model which describes the fragmentation
446: functions of all the octet baryons involves a total of {\it eight}
447: free parameters:
448: 
449: \begin{equation}
450: N_v, \alpha_v, \beta_v, N_s, \alpha_s, \beta_s, b, c.
451: \end{equation}
452: 
453: 
454: 
455: 
456: 
457: For a fit to the experimental data, the fragmentation functions
458: have to be evolved from the initial scale $Q_0$ to the scale of
459: the experiments. We take the input scale $Q_0^2=1.0 ~\rm{GeV}^2$
460: and the QCD scale parameter $\Lambda_{QCD}= 0.3~ \rm{GeV}$, and
461: determine the free parameters of the model by fitting the
462: experimental data~\cite{ALEPH95,ALEPH98,OPAL97b,JPG}
463: on the differential cross sections
464: 
465: 
466: \begin{equation}
467: \frac{1}{\sigma_{tot}}\frac{d \sigma}{d x_E} =\frac{ \sum\limits_q
468: \hat{C}_q \left [ D_q^B (x_E,Q^2)+D_{\bar{q}}^B (x_E,Q^2)
469: \right ]} {\sum\limits_q \hat{C}_q} \label{crosection}
470: \end{equation}
471: for semi-inclusive octet baryon production  $e^+e^- \to B + X$,
472: where $\sigma_{tot}$ is the total cross section for the process
473: and $x_E=2 E_B/\sqrt{s}$. Here  $s$ is the total center-of-mass
474: (c.m.) energy squared, and $E_B$ the energy of the produced proton
475: in the $e^+e^-$ c.m. frame. In (\ref{crosection}), $\hat{C}_q$
476: reads
477: 
478: \begin{equation}
479: \hat{C}_q=e_q^2-2 \chi_1 v_e v_q e_q+ \chi_2 (a_e^2+v_e^2)
480: (a_q^2+v_q^2),\label{hatC}
481: \end{equation}
482: with
483: \begin{equation}
484: \chi_1=\frac{1}{16 \sin^2 \theta_W \cos^2 \theta_W}
485: \frac{s(s-M_Z^2)}{(s-M_Z^2)^2+M_Z^2\Gamma_Z^2},
486: \end{equation}
487: 
488: \begin{equation}
489: \chi_2=\frac{1}{256 \sin^4 \theta_W \cos^4 \theta_W}
490: \frac{s^2}{(s-M_Z^2)^2+M_Z^2\Gamma_Z^2},
491: \end{equation}
492: \begin{equation}
493: a_e=-1,
494: \end{equation}
495: \begin{equation}
496: v_e=-1+4 \sin^2 \theta_W,
497: \end{equation}
498: \begin{equation}
499: a_q=2 T_{3q},
500: \end{equation}
501: and
502: 
503: \begin{equation}
504: v_q=2 T_{3q}-4 e_q \sin^2 \theta_W,
505: \end{equation}
506: where $T_{3q}=1/2$ for $u$, while $T_{3q}=-1/2$ for $d$, $s$
507: quarks, $e_q$ is the charge of the quark in units of the proton
508: charge, $\theta$ is the angle between the outgoing quark and the
509: incoming electron, $\theta_W$ is the Weinberg angle, and $M_Z$ and
510: $\Gamma_Z$ are the mass and width of $Z^0$.
511: 
512: 
513: We perform a leading order (LO) analysis since the results in
514: Refs.~\cite{Flo98b,Kniehl00} show that the leading order fit is of
515: similar quality as the next-to-leading order fit.  Also, the LO
516: analysis should be enough in order to outline the qualitative
517: feature of mass suppression in baryon production. In addition, we
518: only use $z > 0.1$ data samples because understanding the very
519: low-$z$ region data needs further modifications to the evolution
520: of the fragmentation functions~\cite{Flo98b,Kniehl00}. However, we
521: find that some of data in the low-$z$ region can still be
522: described by our fragmentation functions. With the above mentioned
523: cut, we have a total of 157 experimental data
524: points~\cite{ALEPH95,ALEPH98,OPAL97b,JPG}. Eight free parameters
525: of our initial parameterizations are determined by performing a
526: fit to the experimental data. The total $\chi^2$ value of the fit
527: is 192.362, which corresponds to $\chi^2$/point=1.225. The values
528: of the parameters of our model are given in Table~\ref{table1}.
529: 
530: 
531: 
532: 
533: 
534: \begin{table*}
535: \caption{The parameters for the diquark model with the strangeness suppression}
536: \begin{tabular}{|c|c|c|c|c|c|c|c|}
537: \hline
538: %&&\\
539: $N_v$ & $\alpha_v$ & $\beta_v$ & $N_s$ & $\alpha_s$ & $\beta_s$
540: & b ($\rm{GeV}^{-2}$) & c   \\
541: %&&\\
542: \hline\raisebox{0pt}[12pt][6pt] 161.602 & 1.450 & 4.313 & 121.292 & -0.251
543: & 8.161 & 3.394 & 0.241  \\[4pt]
544: \hline
545: \end{tabular}
546: \label{table1}
547: %\end{footnotesize}
548: \end{table*}
549: 
550: 
551: 
552: 
553: \begin{figure*}
554: %\includegraphics{ys1f1.eps}
555: \includegraphics[width=10cm,height=16cm]{ys1f1.eps}
556: \caption[*]{\baselineskip 13pt The comparison of our fit results
557: (thick solid curves) for the $x_E$ dependence of the inclusive
558: octet baryon production cross section $(1/\sigma_{tot})d \sigma/d
559: x_E$ in $e^+e^-$ annihilation and the experimental
560: data~\cite{ALEPH95,ALEPH98,OPAL97b,JPG}.  The thin solid curves
561: correspond to the results with the hyperon fragmentation functions
562: deduced directly from the proton fragmentation functions by using
563: SU(3) symmetry.} \label{ys1f1}
564: \end{figure*}
565: 
566: 
567: 
568: \begin{figure*}
569: %\includegraphics{ys1f2.eps}
570: \includegraphics[width=10cm,height=16cm]{ys1f2.eps}
571: \caption[*]{\baselineskip 13pt The comparison of our fit results
572: for the $x_E$ dependence of the inclusive $\Lambda$
573: production cross section $(1/\sigma_{tot})d \sigma/d
574: x_E$  in $e^+e^-$ annihilation and
575: the experimental data~\cite{JPG,Flo98b}.} \label{ys1f2}
576: \end{figure*}
577: 
578: 
579: 
580: In Fig.~\ref{ys1f1}, we give the fit results (thick solid curves)
581: as compared with the experimental data. In order to provide a
582: clear comparison between the experimental data and the fit curves,
583: only part of data are shown in the figure and a similar fit
584: quality is obtained for other data points. We also present the
585: results for the cross sections for hyperon production with the
586: hyperon fragmentation functions deduced directly from the proton
587: fragmentation functions by means of the SU(3) flavor symmetry
588: relation (see thin solid curves in Fig.~\ref{ys1f1}). By comparing
589: the thick and thin curves, one can find that the mass suppression
590: effect in the cross section of hyperon production is significant.
591: In addition, we also show in Fig.~\ref{ys1f2} the fit results for
592: $\Lambda$ production at various center of mass energies, which
593: indicates that the QCD evolution behavior of the fragmentation
594: functions is reasonable.
595: 
596: 
597: 
598: \section{Octet baryon production in charged lepton DIS}
599: 
600: 
601: \subsection{Unpolarized case}
602: 
603: In the above section, we showed a strong mass suppression effect
604: in hyperon production. This is an extra SU(3) symmetry breaking
605: distinct from that due to the quark and diquark mass differences
606: in the diquark model fragmentation functions, and is effectively
607: described by a mass suppression factor in the octet baryon
608: fragmentation functions. We need a cross check of this mass
609: suppression effect from a different process. Thus we apply the
610: obtained fragmentation functions to calculate the cross sections
611: of octet baryon production in charged lepton DIS.
612: 
613: 
614: To leading order, the cross section for the process
615: 
616: \begin{equation}
617: l + p \to B + X
618: \end{equation}
619: can be expressed as
620: 
621: \begin{equation}
622: \frac{1}{\sigma_{tot}}\frac{{\rm{d}} \sigma}{{\rm{d}} z {\rm{d}} x}
623: = \frac{\sum e_q^2 q(x) D_q^B(z) + (q \to
624: \bar{q})} {\sum e_q^2},
625: \end{equation}
626: where $q(x)$ is the quark distribution in the target nucleon. By
627: inserting the fragmentation functions for the octet baryons into
628: the above cross section, and using the CTEQ5~\cite{CTEQ5} quark
629: distributions in the target nucleon, we get the numerical results
630: shown in Fig.~\ref{ys1f3}, where the $x$-integrated cross sections
631: $(1/\sigma_{tot}){\rm{d}} \sigma / {\rm{d}} z$
632: for baryon (thick solid curves) and anti-baryon (thick dashed
633: curves) production are compared with the available experimental
634: data. In the calculation we have taken $Q^2=50~ \rm{GeV}^2$ and
635: the $x$ integration range [0.02, 0.4].
636: 
637: 
638: \begin{figure*}
639: %\includegraphics{ys1f3.eps}
640: \includegraphics[width=10cm,height=16cm]{ys1f3.eps}
641: \caption[*]{\baselineskip 13pt The cross sections for baryons
642: (thick solid curves) and anti-baryons (thick dashed curves)
643: production in charged lepton DIS, obtained with our fragmentation
644: functions. The results with the hyperon fragmentation functions
645: deduced directly from the proton fragmentation functions by using
646: the SU(3) symmetry are also shown for hyperons (thin solid curves)
647: and anti-hyperons (thin dashed curves) production. The
648: experimental data are taken from
649: Refs.~\cite{NPB480,ZPC61,PLB150}.
650: The original SIDIS data in terms of $x_F$
651: have been converted to the variable $z$ by using the method of
652: Ref.~\cite{Flo98b}.}\label{ys1f3}
653: \end{figure*}
654: 
655: 
656: 
657: We find that the theoretical predictions are compatible with the
658: available experimental data. In Fig.~\ref{ys1f3}, we also show the
659: calculated results with the hyperon fragmentation functions
660: deduced directly from the proton fragmentation functions by using
661: the SU(3) symmetry (thin curves). The experimental data are taken
662: from Refs.~\cite{NPB480,ZPC61,PLB150}. The data points with full
663: circles, triangles and squares are for particle production
664: measured by the E665, EMC and H1 collaborations, respectively;
665: open circles and triangles indicate the data for anti-particle
666: production measured by the E665 and EMC collaborations,
667: respectively. The mass suppression effect in hyperon production,
668: especially in $\Sigma$ and $\Xi$ production, is evident.
669: Therefore, the hyperon production in charged DIS is an ideal place
670: to check the proposed mass suppression effect.
671: 
672: 
673: 
674: \subsection{Polarized case}
675: 
676: Recently experimental data on the spin transfer to $\Lambda$ in
677: charged lepton DIS have become available. The spin transfer is a
678: good observable to check the helicity structure of the
679: fragmentation functions for a baryon. In longitudinally polarized
680: charged lepton DIS on an unpolarized  proton target, the produced
681: baryon polarization along its own momentum axis is given in the
682: quark parton model by
683: \begin{equation}
684: P_{B}(x,y,z) = P_b D(y)A_{B}(x,z)~,
685: \label{PL}
686: \end{equation}
687: where $P_b$ is the polarization of the charged lepton beam, $D(y)$
688: with $y=\nu/E$ is  the longitudinal depolarization factor of the
689: virtual photon with respect to the parent lepton, and
690: \begin{equation}
691: A_{B}(x,z)= \frac{\sum\limits_{q} e_q^2 [q(x,Q^2) \Delta
692: D_q^B(z,Q^2) + ( q \rightarrow \bar q)]}
693: {\sum\limits_{q} e_q^2 [q (x,Q^2)
694: D^B_q(z,Q^2) + ( q \rightarrow \bar q)]}~,
695: \label{DL}
696: \end{equation}
697: is the longitudinal spin transfer to the baryon $B$.
698: 
699: In order to check the spin structure of the obtained fragmentation
700: functions, we calculate the $x$-integrated spin transfer to octet
701: baryons in charged lepton DIS. The numerical results are shown in
702: Fig.~\ref{ys1f4}. Our theoretical predictions are consistent with
703: the available experimental data on $\Lambda$ production.
704: 
705: 
706: \begin{figure*}
707: %\includegraphics{ys1f4.eps}
708: \includegraphics[width=10cm,height=16cm]{ys1f4.eps}
709: \caption[*]{\baselineskip 13pt The spin transfer to
710: baryons (solid curves) and anti-baryons
711: (dashed curves) production
712: in charged lepton DIS obtained with our
713: fragmentation functions.
714: The experimental data are taken from
715: Refs.~\cite{HERMES,E665}.}\label{ys1f4}
716: \end{figure*}
717: 
718: 
719: \section{Octet baryon production in $pp$ collisions}
720: 
721: In the near future, new experimental data will become available on
722: hadron production in $pp$ collisions at BNL-RIHC~\cite{Saito}.
723: Therefore, it is interesting to predict the cross sections for octet baryon
724: production in $pp$ collisions in order to have a further check of
725: the mass suppression effect in octet hyperon production.
726: 
727: In leading order of perturbative QCD, the differential
728: cross section for the $p p \to B X$ process can be
729: schematically written in a factorized form as~\cite{MSSY9}
730: 
731: \begin{multline}
732: E {d^3\sigma \over d^3p}= \sum \limits_{abcd}
733: \int_{\bar x_a}^1 dx_a
734: \int_{\bar x_b}^1 dx_b f_a^{\tilde{A}}(x_a,Q^2)f_b^{\tilde{B}}(x_b,Q^2)
735: D_c^B(z,Q^2) {1 \over \pi z}{ d\hat \sigma \over d\hat t}(ab \to cd)~,
736: \label{pp1}
737: \end{multline}
738: with
739: \begin{equation}
740: \bar x_a={x_{T}e^y \over 2-x_Te^{-y}}\ ,\
741: \bar x_b={x_ax_Te^{-y}\over 2x_a-x_Te^y}\ ,\  z={x_T\over 2x_b} e^{-y}
742: + {x_T \over 2x_a}e^y \ ,
743: \label{xz}
744: \end{equation}
745: where $x_T=2p_T/\sqrt s$, $\sqrt s$ is the center of mass energy
746: of the $pp$ collision; $p_T$, $E$ and $y$ are the transverse
747: momentum, energy and rapidity of the produced baryon $B$;
748: $f_a^{\tilde{A}} (x_a,Q^2)$ and $ f^{\tilde{B}}_b (x_b,Q^2)$ are
749: the unpolarized distribution functions of partons $a$ and $b$ in
750: the protons ${\tilde{A}}$ and ${\tilde{B}}$ at the scale
751: $Q^2=p_T^2$; $D_c^B( z, Q^2)$ is the fragmentation function which
752: we have obtained in Sec. II; $\frac{d \hat{\sigma}}{d \hat{t}}$ is
753: the differential cross section for the sub-process $a+b \to c +
754: d$, and  $\hat{t}=-x_a p_T \sqrt{s} e ^ {-y} /z$ is the Mandelstam
755: variable at the parton level.
756: 
757: By charge-conjugation invariance, the $e^+e^- \to B X$ cross
758: section for baryon production should be equal to that for the
759: corresponding antibaryon production process. Therefore, only the
760: combinations $D^B_q + D^{\bar{B}}_q$ can be determined, and the
761: same holds for the antiquark fragmentation functions. However, in
762: $pp$ collisions we can observe differences in the cross sections
763: for baryon and anti-baryon production. Therefore in this case we
764: also predict the cross sections for anti-baryon $\bar{B}$
765: production, whose quark fragmentation functions can be obtained
766: according to the matter-antimatter symmetry $D_{q,\bar{q}}^{B} (z)
767: =D _{\bar{q},q}^{\bar{B}}(z)$.
768: 
769: 
770: By adopting the LO set of unpolarized parton distributions of
771: Ref.~\cite{GRV95}, we present in Fig.~\ref{ys1f5} the cross
772: sections for octet baryons (thick solid curves) and antibaryons
773: (thick dashed curves) produced in $pp$ collisions. These results
774: are calculated at $\sqrt{s}=500~\rm{GeV}$ and $p_T=
775: 15~\rm{GeV/c}$. As a comparison, we also calculate the cross
776: sections with the hyperon fragmentation functions deduced directly
777: from the proton fragmentation functions by using the SU(3)
778: symmetry relation (thin curves). By comparing the thick and thin
779: curves, one can find that the mass suppression effect in hyperon
780: production from $pp$ collisions is also significant. Therefore,
781: the cross sections for the octet hyperon production in $pp$
782: collisions should be another ideal domain where the mass
783: suppression effect can be checked. Although some experiments for
784: baryon production in $pp$ collisions have been done~\cite{pptob},
785: the available data were taken in the low-$p_T$ region. We need
786: some data at high-$p_T$ in order to check our partonic framework
787: predictions. This may be realized by
788: RHIC-BNL~\cite{Saito,MSSY9,Bunce} in the near future.
789: 
790: 
791: 
792: \begin{figure*}
793: %\includegraphics{ys1f5.eps}
794: \includegraphics[width=10cm,height=16cm]{ys1f5.eps}
795: \caption[*]{\baselineskip 13pt The cross sections for baryons
796: (thick solid curves) and anti-baryons (thick dashed curves)
797: production in $pp$ collisions are predicted at $\sqrt{s}= 500~
798: \rm{GeV}$ and $p_T= 15~ \rm{GeV/c}$. The results with the hyperon
799: fragmentation functions deduced directly from the proton
800: fragmentation functions by using the SU(3) symmetry relation are
801: also shown for hyperons (thin solid curves) and anti-hyperons
802: (thin dashed curves) production. }\label{ys1f5}
803: \end{figure*}
804: 
805: 
806: 
807: \section{Summary}
808: 
809: Based on the quark diquark model, the fragmentation functions for
810: all octet baryons are related by the SU(3) relation. Nevertheless
811: the hadronic multiplicities measurements in electron-positron
812: annihilation indicate a strong suppression in octet hyperon
813: production as compared with proton production, which cannot be
814: explained by the SU(3) symmetry breaking within the diquark model
815: framework. Inspired by the phenomenology of the string model, we
816: proposed an overall mass suppression factor for both unpolarized
817: and polarized octet baryon fragmentation functions, retaining the
818: flavor and spin structure of the fragmentation functions given by
819: the diquark model. We found that the diquark model with the mass
820: suppression factor can be used to describe quite accurately the
821: fragmentation functions for all octet baryons with {\it eight}
822: free parameters. The parameters were determined by a fit to the
823: available experimental data on the octet baryon production in
824: electron-positron annihilation. We used eight parameters; three
825: for the unpolarized valence down quark to proton fragmentation
826: function $D_{d_v}^p (z)$ (see Eq.~(\ref{fit1})), while all other
827: unpolarized and polarized valence quark fragmentation functions
828: for the proton follow from the diquark model (Eqs.~(\ref{eq16})-
829: (\ref{eq20})); three parameters for the sea quark fragmentation
830: functions (Eq.~(\ref{fit2})); and finally two more for the
831: suppression factor (Eq.~(\ref{ssp})). In addition, the diquark
832: model plays an important role in relating fragmentation functions
833: for all octet baryons to each other (see Eq.~(\ref{su3})).
834: 
835: 
836: The mass suppression factor leads to an enormous simplification in
837: our analysis and plays an important role in our understanding of
838: the experimental data on the unpolarized hyperon production in
839: $e^+e^-$ annihilation. This mechanism needs to be further checked.
840: The octet baryon fragmentation functions, determined from the
841: $e^+e^- \to B X$ process, can be used to predict inclusive single
842: baryon production cross sections in other processes, like $pp$,
843: $p\bar{p}$, $ep$, $\nu p$, $\mu p$ and $\gamma p$ scattering. With
844: the obtained fragmentation functions, we calculated the cross
845: section for octet baryon production in charged lepton DIS, and our
846: predictions are compatible with the available experimental data.
847: Furthermore, we predicted cross sections for octet baryon
848: production in $pp$ collisions. We investigated the mass
849: suppression effect of hyperon production in charged lepton DIS and
850: $pp$ collisions, and found that these two processes are ideal
851: places for checking the proposed mass suppression effect when
852: further experimental data become available.
853: 
854: 
855: 
856: 
857: \begin{acknowledgments}
858: 
859: We would like to thank A. Sch\"{a}fer and J. Soffer for valuable
860: discussions and comments. This work is supported by the Alexander von Humboldt
861: Stiftung Foundation, and partially supported by the National
862: Natural Science Foundation of China under Grant Number 10175074,
863: by the Foundation for University Key Teacher, by the Ministry of
864: Education (China), by the DFG and BMBF, and by Fondecyt (Chile)
865: project 1030355.
866: 
867: \end{acknowledgments}
868: 
869: 
870: 
871: \begin{thebibliography}{99}
872: 
873: %1
874: \bibitem{Lattice}
875: M. G\"ockeler, R. Horsley, D. Pleiter, P. E. L. Rakow, S. Sch\"afer,
876: A. Sch\"afer, and G. Schierholz, Phys. Lett. B {\bf 545}, 112 (2002).
877: 
878: %2
879: \bibitem{GLR}
880: V.N.~Gribov and L.N.~Lipatov, Phys. Lett. {\bf B 37}, 78 (1971);
881: Sov. J. Nucl. Phys. {\bf 15}, 675 (1972).
882: 
883: %3
884: \bibitem{Bro97}
885: S.J.~Brodsky and B.-Q.~Ma, Phys. Lett. {\bf B 392}, 452 (1997).
886: 
887: %4
888: \bibitem{MSSY}
889: B. Q. Ma, I. Schmidt, J. Soffer, J. J. Yang, Phys. Lett. {\bf B 547},
890: 245 (2002); Eur. Phys. J. {\bf C 16}, 657 (2002); Phys. Rev.
891: {\bf D 65}, 034004 (2002); Phys. Rev. {\bf D 64}, 014017
892: (2001), Erratum-{\it ibid}, {\bf D 64}, 099901 (2001);
893: Phys. Lett. {\bf B 488}, 254 (2000); Phys. Rev. {\bf D 62},
894: 114009 (2000).
895: 
896: %5
897: \bibitem{MSY}
898: B. Q. Ma, I. Schmidt, J. J. Yang, Phys. Lett. {\bf B 477},
899: 107 (2000); Phys. Rev. {\bf D 62}, 114009 (2000).
900: 
901: 
902: %6
903: \bibitem{CU95}
904: P. V. Chliapnikov, V. A. Uvarov, Phys. Lett. B {\bf 345}, 313 (1995).
905: 
906: %7
907: \bibitem{Barger}
908: V. D. Barger and R. J. N. Phillips, {\it{Collider Physics}}
909: (Addison-Wesley, Reading, MA, 1987).
910: 
911: %8
912: \bibitem{Nza95}
913: R.~Jakob, P.J.~Mulders, J.~Rodrigues, Nucl. Phys. A {\bf 626},
914: 937 (1997);
915: 
916: M.~Nzar and P.~Hoodbhoy, Phys. Rev. D {\bf 51}, 32 (1995).
917: 
918: %9
919: \bibitem{a01}
920: J. J. Yang, Phys. Rev. D {\bf 64}, 074010 (2001);
921: Nucl. Phys. A {\bf 699}, 562 (2002).
922: 
923: 
924: %10
925: \bibitem{a06}
926: J. J. Yang, Phys. Rev. D {\bf 65}, 094035 (2002).
927: 
928: 
929: %11
930: \bibitem{ALEPH96}
931: ALEPH Collaboration, D.~Buskulic {\it et al.}, Phys. Lett.  B
932: {\bf 374}, 319 (1996).
933: 
934: 
935: %12
936: \bibitem{DELPHI95}
937: DELPHI Collaboration, Report No.DELPHI 95-86 PHYS 521,
938: CERN-PPE-95-172, presented at the EPS-HEP 95 conference, Brussels,
939: 1995.
940: 
941: %13
942: \bibitem{OPAL97}
943: OPAL Collaboration, K.~Ackerstaff {\it et al.},
944: Eur. Phys. J.  C {\bf 2}, 49 (1998).
945: 
946: %14
947: \bibitem{HERMES}
948: HERMES Collaboration, A.~Airapetian {\it et al.},
949: Phys. Rev. D {\bf 64}, 112005 (2001).
950: 
951: %15
952: \bibitem{E665}
953: E665 Collaboration, M.R.~Adams {\it et al.}, Eur.
954: Phys. J. C {\bf{ 17}}, 263 (2000).
955: 
956: %16
957: \bibitem{NOMAD}
958: NOMAD Collaboration: P. Astier {\it et al.}, Nucl. Phys. B {\bf 588}, 3 (2000),
959: CERN-EP/2000-111; Nucl. Phys. B {\bf 605}, 3 (2000), CERN-EP/2001-028.
960: 
961: %17
962: \bibitem{PDG00}
963: Particle Data Group, D. E. Groom {\it et al.}, Eur. Phys. J.
964: C {\bf 15}, 1 (2000).
965: 
966: %18
967: \bibitem{Rastogi}
968: A. Rastogi, Phys. Rev. D {\bf 59}, 114012 (1999);
969: 
970: D. Indumathi, H.S. Mani, and Anubha Rastogi,
971: Phys. Rev. D {\bf 58}, 094014 (1998).
972: 
973: %19
974: \bibitem{String}
975: S. B. Chun, C. D. Buchanan, Phys. Lett. B 308 (1993) 153.
976: 
977: 
978: %20
979: \bibitem{ALEPH95}
980: ALEPH Collaboration, D. Buskulic {\it et al.}, Z. Phys. C {\bf 66}, 355 (1995).
981: 
982: %21
983: \bibitem{ALEPH98}
984: ALEPH Collaboration, R. Barate {\it et al.},  Phys. Rep. {\bf 294}, 1 (1998).
985: 
986: 
987: %22
988: \bibitem{OPAL97b}
989: OPAL Collaboration, G. Alexander {\it et al.}, Z. Phys. C {\bf 73}, 587 (1997).
990: 
991: 
992: %23
993: \bibitem{JPG}
994: G. D. Lafferty, P. I. Reeves, and M. R. Whalley, J. Phys. G {\bf
995: 21}, A1 (1995).
996: 
997: 
998: %24
999: \bibitem{Flo98b}
1000: D.de Florian, M.~Stratmann, and W.~Vogelsang, Phys. Rev. D {\bf
1001: 57}, 5811 (1998).
1002: 
1003: 
1004: %25
1005: \bibitem{Kniehl00}
1006: B. A. Kniehl, G. Kramer, B. P\"{o}tter, Nucl. Phys. B {\bf 582}, 514 (2000).
1007: 
1008: %26
1009: \bibitem{CTEQ5}
1010: CTEQ Collaboration, H.L.~Lai {\it et al.}, Eur. Phys. J. C {\bf 12},
1011: 375 (2000).
1012: 
1013: %27
1014: \bibitem{NPB480}
1015: H1 Collaboration, S. Aid {\it et al.}, Nucl. Phys. B {\bf 480}, 3 (1996).
1016: 
1017: %28
1018: \bibitem{ZPC61}
1019: E665 Collaboration, M. R. Adams {\it et al.}, Z. Phys. C {\bf 61}, 539 (1994).
1020: 
1021: %29
1022: \bibitem{PLB150}
1023: EMC Collaboration, M. Arneodo {\it et al.}, Phys. Lett. B {\bf 150},
1024: 458 (1985).
1025: 
1026: 
1027: %30
1028: \bibitem{Saito}
1029: N.~Saito, Plenary talk in 14th International Spin Physics
1030: Symposium (SPIN 2000),  Osaka, Japan, 16-21 Oct 2000.
1031: 
1032: 
1033: 
1034: %31
1035: \bibitem{MSSY9}
1036: B.Q.~Ma, I.~Schmidt, J.~Soffer, and J.J.~Yang,
1037: Nucl. Phys. A {\bf 703}, 346 (2002).
1038: 
1039: 
1040: %32
1041: \bibitem{GRV95}
1042: M. Gl\"{u}ck, E. Reya, and A. Vogt, Z. Phys. C {\bf 67},
1043: 433 (1995).
1044: 
1045: %33
1046: \bibitem{pptob}
1047: LEBC-EHS Collaboration, M. Aguilar-Benitez {\it et al.}, Z. Phys. C {\bf 50},
1048: 405 (1991);
1049: 
1050: I. Derado, K. Kadija, M. Malecki, N. Schmitz, P. Seyboth,
1051: Z. Phys. C {\bf 50}, 31 (1991);
1052: 
1053: M. Gazdzicki, Ole Hansen, Nucl. Phys. A {\bf 528}, 754 (1991).
1054: 
1055: 
1056: 
1057: %34
1058: \bibitem{Bunce} G.~Bunce, N.~Saito, J.~Soffer and W.~Vogelsang,
1059: Ann. Rev. Nucl. Part. Scie. {\bf 50}, 525 (2000).
1060: 
1061: 
1062: \nonfrenchspacing
1063: \end{thebibliography}
1064: \end{document}
1065: \nonfrenchspacing
1066: \end{thebibliography}
1067: