1: % Dear Valera,
2: % I have included the deuteron Table in our Text
3: % (and corrected the Table 1-5 captions)
4: % Best wishes, Misha
5: %------------------------------------------
6: \documentstyle[12pt,epsfig]{article}
7: \begin{document}
8:
9:
10:
11: \title
12: %{EMC effect: role of the boundness and motion of nucleons in
13: %nuclei}
14: {Contribution of boundness and motion of nucleons to the EMC effect}
15: \author{
16: B.L. Birbrair, M.G. Ryskin and V.I. Ryazanov\\
17: Petersburg Nuclear Physics Institute\\
18: Gatchina, St. Petersburg 188300, Russia}
19: \date{}
20: \maketitle
21:
22:
23:
24: \begin{abstract}
25:
26: The kinematical corrections to the structure function of nucleon in
27: nucleus due to the boundness and motion of nucleons arise from
28: the
29: excitation of the doorway states for one-nucleon transfer reactions
30: in
31: the deep inelastic scattering on nuclei.
32:
33: \end{abstract}
34:
35: \section{Introduction}
36:
37: It is known more than 20 years that the cross section of deep
38: inelastic
39: scattering (DIS) on nuclear target is not equal to the sum of cross
40: sections on free nucleons \cite{EMC}. This means that the
41: interaction
42: inside the nucleus distorts the parton distribution in a nucleon.
43: But at first it is necessary to single out the kinematical effects arising
44: from the boundness and motion of nucleons in nuclei because otherwise it is
45: hardly possible to conclude what actually happends with the nucleon structure
46: functions in nuclear matter. This is the aim of our work rather than the
47: explanation of the EMC effect. The above kinematical effect is due to
48: % Part of the effect is of a pure kinematical nature because of
49: the fact
50: that the four-momentum of nucleon in nucleus is not equal to that
51: of a free nucleon. Indeed,
52: the heavy photon ($\gamma^*$) is absorbed by
53: a single nucleon and the Deep Inelastic Scattering (DIS) proceeds via
54: the following stage:
55: $$l\ +\ A\rightarrow l'\ +\ X\ +\ (A-1)^*\; ,$$
56: where $l$ and $l'$ denote the incoming and outgoing leptons, $X$ - the
57: final hadronic state after the $\gamma^*$-nucleon interaction and $A$
58: is the target nucleus. Before absorbing the heavy photon
59: $(\gamma^*$) the struck nucleon has a certain energy-momentum
60: distribution in nucleus. Besides this the ``residual'' ($A-1$) nucleus
61: is excited.\\
62:
63: There were a few attempts to account for the Fermi motion,
64: boundness
65: and the change of the effective $\gamma$-nucleon flux factor inside
66: the
67: nucleus \cite{BR,Vagr,FS} (see \cite{Arn} and references therein for
68: more details). They all were based on a seemingly obvious
69: assumption
70: that the energy-momentum distribution of the struck nucleon is
71: described by the ground-state one-nucleon spectral function.
72: \begin{equation}
73: S_g({\bf p},\varepsilon)=\langle A_0|a^+({\bf p})\delta\bigg(
74: \varepsilon+H-{\cal E}_0(A)\bigg)a({\bf p})|A_0\rangle\ ,
75: \end{equation}
76: where $|A_0\rangle$ is the ground state of target nucleus $A$,
77: $a({\bf
78: p} )$ and $a^+({\bf p})$ are operators of nucleon with the momentum
79: ${\bf p}$ (the spin and isospin variables are omitted), $H$ is the
80: nuclear Hamiltonian in the second quantization and
81: ${\cal E}_0(A)$
82: is the ground-state binding energy of nucleus $A$. The calculations
83: \cite{Vagr} were performed by using the following semiempirical
84: model
85: for the quantity (1): the nucleon energy distributions were described
86: by the experimental data on the separation energies of protons
87: from the
88: $(e,e'p)$ reactions \cite{6} (the difference between the proton and
89: neutron separation energies was neglected leading to about 10\%
90: error)
91: and calculating the momentum distributions within the harmonic
92: oscillator model with the parameter
93: $\hbar\omega_0=(45\,A^{-1/3}-25\,A^{-2/3})$ MeV reproducing the
94: observed rms
95: radii of nuclei.
96:
97: However nobody realized in this connection that the DIS on nuclei is
98: rapid process, and therefore the energy-momentum distribution of
99: struck
100: nucleon is described by the spectral function of nuclear mode
101: which is
102: excited via a sudden perturbation rather than that of the ground
103: state.
104: Our work is based on the fact that the relevant mode is provided by
105: the
106: doorway states for the one-nucleon transfer reactions. As
107: demonstrated
108: in \cite{3}--\cite{BR3} the above states are eigenstates of
109: nucleon
110: in the static nuclear field.
111:
112: Recall that the microscopic nuclear models are based on certain
113: approximations for the in-medium nucleon mass operator $M$. For
114: instance the nuclear shell model potential is the approximation for
115: the
116: mass operator at nuclear Fermi-surface, the optical model potential
117: is
118: dealing with the mass operator at low and intermediate energies,
119: etc.
120: In all the available approaches
121: %both cases
122: the mass operator includes all the Feynman
123: diagrams which
124: are irreducible in the one-particle channel, and therefore it cannot be
125: calculated. Instead it is described by a set of phenomenological
126: parameters.
127:
128: In contrast to the above models the nuclear static potential is the
129: mass operator at the infinite value of the energy variable. Only the
130: Hartree diagrams with the free space ({\em i.e.} vacuum) nuclear
131: forces
132: survive in this case thus permitting the model-independent
133: calculation
134: of the static field. So the doorway states (DS) under consideration
135: appear to be the unique nuclear object, both model-independent and
136: described by the exactly soluble problem.
137:
138: The calculation \cite{BR2} showed that the rms radii of the DS
139: density
140: distributions are appreciably less than those of the ground-state
141: ones:
142: for instance, the value of $\langle
143: r^2\rangle=A^{-1}\int\rho(r)r^2d^3r$ is $\rm10.88\,fm^2$ for the
144: ground
145: state of $^{40}$Ca being only $\rm8.76\,fm^2$ for the DS. As a
146: result
147: the nucleon motion ({\em i.e.} the value of $\langle{\bf p}^2\rangle$)
148: was underestimated in \cite{Vagr} by about 25\% \footnote{By the same
149: reason the kinematical effect was underestimated in Ref.\cite{BLev} as
150: well.}.\\
151:
152:
153: In Sec. 2 we briefly describe the formalism of the DS. In Sec. 3 the
154: DIS structure functions $F_2$ are calculated for different nuclei and
155: deuteron; to single out the boundness and motion effects we
156: disregarded
157: the possible changes of the parton distribution inside the nucleon in
158: nucleus. In the last Sec. 4 we compare the calculated ratios
159: $2F_{2A}/AF_{2D}$ with the available experiments.
160:
161:
162:
163:
164: %In sect.2 we recall the formalism of ``doorway'' states and describe
165: %the way used to calculate eigen functions. In sect.3 the DIS structure
166: %function $F_2$ is calculate for different nuclei and for deuteron
167: %assuming the parton distribution inside the nucleon in nucleus to be
168: %the same as that for the free nucleon. In the last sect.4 we compare
169: %the calculated ratio $2F_{2A}/AF_{2D}$ with the avaliable experiments.
170:
171: \section{Doorway states for one-nucleon transfer reactions}
172:
173: \subsection{Theory}
174:
175: Evolution of the state arising from the one-nucleon transfer to the
176: nuclear ground state $|A_0\rangle$ at the initial time moment $t=0$ is
177: described by the single-particle propagator \cite{7}
178: \begin{eqnarray}
179: && S(x,x';\tau)\ =\ -i\langle A_0|T\psi(x,\tau)\psi^\dag(x',0)|A_0
180: \rangle\ = \nonumber\\
181: = && i\theta(-\tau)\sum^{(A-1)}_j
182: \Psi_j(x)\Psi^\dag_j(x')e^{-iE_j\tau}-i\theta(\tau) \sum^{(A+1)}_k
183: \Psi_k(x)\Psi^\dag_k(x')e^{-iE_k\tau}.
184: \end{eqnarray}
185: At $\tau<0$ it describes the evolution of the hole state,
186: \begin{equation}
187: \Psi_j(x)=\langle(A-1)_j|\psi(x)|A_0\rangle\ ,\quad E_j=
188: {\cal E}_0(A)-{\cal E}_j(A-1)\ ,
189: \end{equation}
190: when the nucleon is removed from the ground state $A_0$, whereas at
191: $\tau>0$ the evolution of the particle state is described,
192: \begin{equation}
193: \Psi_k(x)=\langle A_0|\psi(x)|(A+1)_k\rangle\ , \quad E_k=
194: {\cal E}_k(A+1)-{\cal E}_0(A)\ ,
195: \end{equation}
196: when nucleon is added to the ground state $A_0$. The quantities
197: ${\cal E}_j(A-1)$, ${\cal E}_k(A+1)$ and ${\cal E}_0(A)$ are total
198: binding energies of the states $|(A-1)_j\rangle$ of the $(A-1)$
199: nucleus, the states $|(A+1)_k\rangle$ of the $(A+1)$ nucleus and the
200: ground state $|A_0\rangle$ of the $A$ one.
201:
202: The Fourier transform of the propagator
203: \begin{equation}
204: G(x,x';\varepsilon)=\int S(x,x';\tau)e^{i\varepsilon\tau} d\tau =
205: \sum^{(A-1)}_j\frac{\Psi_j(x)\Psi^\dag_j(x')}{\varepsilon-E_j-i\delta}
206: +\sum^{(A+1)}_k\frac{\Psi_k(x)\Psi^\dag_k(x')}{\varepsilon-E_k+i\delta}\
207: ,
208: %\quad \delta\to\ +0
209: \end{equation}
210: $\quad\hspace{10cm} \delta\to\ +0$\\
211: obeys the Dyson equation
212: \begin{equation}
213: \varepsilon G(x,x';\varepsilon)\ =\ \delta(x-x')+\hat k_xG(x,x';
214: \varepsilon)+ \int M(x,x_1;\varepsilon)G(x_1,x';\varepsilon)dx_1\ ,
215: \end{equation}
216: where $\hat k_x$ is the kinetic energy and the mass operator $M(x,x';
217: \varepsilon)$ includes all Feynman diagrams which are irreducible in
218: the one-particle channel.
219:
220: We are interested in the very beginning of the evolution, i.e. the
221: $\tau\to0$ limit. According to the time-energy Heisenberg relation this
222: is equivalent to the limit $\varepsilon\to\infty$. In this limit
223: \begin{equation}
224: G(x,x';\varepsilon)\ =\ \frac{I_0(x,x')}\varepsilon +\frac{I_1(x,x')}{
225: \varepsilon^2}+\frac{I_2(x,x')}{\varepsilon^3}+\cdots\ ,
226: \end{equation}
227: where (see the definition (1) of the propagator)
228: \begin{equation}
229: I_0(x,x')\ =\ \sum^{(A-1)}_j\Psi_j(x)\Psi_j^\dag(x')+\sum^{(A+1)}_k
230: \Psi_k(x)\Psi^\dag_k(x')=i\bigg[S(x,x';+0)-S(x,x';-0)\bigg];
231: \end{equation}
232: \begin{eqnarray}
233: I_1(x,x')&&=\ \sum^{(A-1)}_jE_j\Psi_j(x)\Psi_j^\dag(x')+\sum^{(A+1)}_k
234: E_k\Psi_k(x)\Psi^\dag_k(x')\ =\nonumber\\
235: &&=\ -\bigg[\dot
236: S(x,x';+0)-\dot S(x,x';-0)\bigg];
237: \end{eqnarray}
238: \begin{eqnarray}
239: I_2(x,x')&&=\
240: \sum^{(A-1)}_jE^2_j\Psi_j(x)\Psi_j^\dag(x')+\sum^{(A+1)}_k
241: E^2_k\Psi_k(x)\Psi^\dag_k(x')\ =\nonumber\\
242: &&=\ -i\bigg[\ddot S(x,x';+0)-\ddot
243: S(x,x';-0)\bigg]
244: \end{eqnarray}
245: the quantities $I_0, I_1$ and $I_2$ thus describing the very beginning
246: of the evolution $\left(\dot S=\frac{\partial S}{\partial\tau}, \ddot S
247: =\frac{\partial^2S}{\partial\tau^2}\right)$.
248:
249: Now consider the mass operator $M(x,x';\varepsilon)$. It includes the
250: energy-independent Hartree diagrams $U_{st}(x)\delta(x-x')$ (which were
251: shown in Fig.3 of Ref.\cite{BR3}) the higher-order diagrams describing
252: the nuclear correlation effects (the lowest-order diagram of such kind
253: was shown in Fig.4a of Ref.\cite{BR3}) and the Fock ones (Fig. 4b of
254: Ref.\cite{BR3}). The correlation diagrams include the propagators of
255: intermediate states thus behaving as $\varepsilon^{-1}$ in the
256: $\varepsilon\to\infty$ limit (see Ref.\cite{8} for more stringent
257: demonstration). The same is valid for the Fock diagrams.
258: %, Fig. 4b.
259: Indeed, the interaction between baryons proceeds via the exchange by
260: some particles (they are quark--antiquark pairs and/or gluons in the
261: QCD) and therefore both the momentum and the energy are transferred
262: through the interaction. As a result the Fock diagrams also include the
263: intermediate state propagators thus being of order of
264: $\varepsilon^{-1}$ in the $\varepsilon\to\infty$ limit. (In Ref.
265: \cite{3} this is demonstrated for the meson-nucleon intermediate
266: state). So the mass operator in this limit is \begin{eqnarray} &&
267: M(x,x';\varepsilon)\ =\ U_{st}(x)\delta(x-x')+ \frac{\Pi(x,x')
268: }\varepsilon\ +\ \cdots \\ &&\ \varepsilon\to\infty \nonumber
269: \end{eqnarray} Introducing the static Hamiltonian \begin{equation}
270: h_{st}\ =\ \hat k_x+U_{st}(x) \end{equation} let us write down the
271: high-energy limit Dyson equation in the form \begin{equation}
272: \varepsilon G(x,x';\varepsilon)=\delta(x-x')+h_{st}G(x,x';\varepsilon)
273: +\int\left(\frac{\Pi(x,x_1)}\varepsilon+\cdots\right)
274: G(x_1,x';\varepsilon)dx_1\ .
275: \end{equation}
276: Putting into (13) the asymptotics (7) and equating coefficients at the
277: same powers of $\varepsilon^{-1}$ we get
278: \begin{eqnarray}
279: && \sum^{(A-1)}_j\Psi_j(x)\Psi^\dag_j(x')+\sum^{(A+1)}_k\Psi_k(x)
280: \Psi^\dag_k(x')\ =\ \delta(x-x')\\
281: && \sum^{(A-1)}_jE_j\Psi_j(x)\Psi^\dag_j(x')+\sum^{(A+1)}_k E_k
282: \Psi_k(x)\Psi^\dag_k(x')\ =\ h_{st}\delta(x-x')
283: %\\
284: %&&
285: \end{eqnarray}
286: \begin{equation}
287: \sum^{(A-1)}_jE^2_j\Psi_j(x)\Psi^\dag_j(x')+\sum^{(A+1)}_k E^2_k
288: \Psi_k(x)\Psi^\dag_k(x')\ =\ h^2_{st}\delta(x-x')+\Pi(x,x').
289: %\end{eqnarray}
290: \end{equation}
291: Equations (9), (12) and (15) may be written as
292: \begin{equation}
293: -\left[\dot S(x,x';+0)-\dot S(x,x';-0)\right]\ =\ h_{st}\delta(x-x')\
294: =\ \left[k_x^{}+U_{st}(x)\right]\delta(x-x')\ .
295: \end{equation}
296: As follows from the lhs of (17) the hamiltonian $h_{st}$ describes the
297: very beginning of the one-nucleon transfer process the eigenstates of
298: $h_{st}$ thus being the doorway states for one-nucleon transfer
299: reactions. On the other hand the rhs of (17) shows that the hamiltonian
300: $h_{st}$ describes the motion of nucleon in nuclear static field
301: $U_{st}(x)$. Indeed, the latter is expressed through the free-space
302: $NN$ forces rather than the effective ones thus being the nucleon field
303: rather than the quasiparticle one. So we proved that the doorway states
304: for one-nucleon transfer fast reactions
305: %on one nucleon
306: are the
307: eigenstates of nucleon in nuclear static field.
308:
309: \subsection{Doorway eigenfunctions}
310: Since the doorway states (DS) describe the motion of the nucleon in
311: nuclear {\it static} field the corresponding eigenfunctions may be
312: calculated in a model independent way. Indeed, the two-particle forces
313: are determined from the experimental data on the elastic
314: nucleon-nucleon scattering
315: (i.e. from the phase shifts analysis)\cite{6b} and the deuteron
316: properties. The necessary information about the multiparticle forces is
317: obtained from the observed energy spectra of the doorway states
318: \cite{3}. So the only additional information needed for
319: calculation of the static field in a given nucleus is that on the
320: nucleon density distributions in this nucleus. In all the nuclei which
321: are treated in the present paper these distributions are spherically
322: symmetric thus leading to the static field with the same symmetry.
323: Hence the quantum mechanical problem is the motion of a particle in a
324: central field. This problem is solved with any desired accuracy and
325: without any simplifications.\\
326:
327:
328: We have to emphasize that the doorway states are not
329: the eigenfunctions of the
330: %full
331: total nuclear Hamiltonian thus being
332: fragmented over the actual nuclear states owing to the
333: correlation effects. The observed spreading width of the DS is
334: about 20 MeV; that is the relaxation time $\sim 0.3\cdot 10^{-22}$
335: sec. This is much larger than the time chracteristic for DIS
336: which is of the order of $2q_0/Q^2 \simeq 1/mx \sim 3\cdot 10^{-24}$
337: sec in the nucleus rest frame. So during the DIS process the DS do
338: not have time to be distorted by the correlations thus permitting the
339: exact account for the nucleon boundness and motion to the EMC effect.\\
340:
341: The relevant energy-momentum distribution of nucleons for DIS is
342: determined by the spectral function of the DS (rather than the ground
343: state one):
344: \begin{equation}
345: S_{DS}(\varepsilon,\vec p)\; =\; S_{p}(\varepsilon,\vec p)\; +\;
346: S_{n}(\varepsilon,\vec p)\; ,
347: \end{equation}
348: where the proton spectral function is
349: \begin{equation}
350: S_{p}(\varepsilon,\vec p)\; =\;\frac 1{4\pi}\sum^{(p)}_\lambda
351: \nu_\lambda f_\lambda(\vec p)\delta(\varepsilon-\varepsilon_\lambda)\;
352: .
353: \end{equation}
354: The sum in the r.h.s. runs over the proton DS,
355: $\lambda$ stands for the angular momentum $j$ and other quantum numbers
356: of a particle state in central field, $\nu_\lambda$ equal to $2j+1$ for
357: the filled states and the actual number of nucleons on partly filled
358: ones, $\varepsilon_\lambda$ are the DS energies and $f_\lambda(p)$ are
359: found by solving the Dirac equation (see Ref.s\cite{3,BR3} for
360: details). \begin{equation} h_{st}\psi_\lambda(\vec
361: r)\;=\;\varepsilon_\lambda\psi_\lambda(\vec r)\; . \end{equation} The
362: function $f_\lambda(p)=u^2_\lambda(p)+w^2_\lambda(p)$, given by the sum
363: of the upper and lower components square of the bi-spinor
364: $\psi_\lambda(p)$ (in momentum space), is normalized by the condition
365: \begin{equation} \int f_\lambda(p)p^2dp=1\; . \end{equation}
366:
367: The neutron spectral function obeys the same relation in which the
368: proton DS are substituted by the neutron ones.\\
369:
370: It is instructive to mention that the spectral functions
371: $S_{DS}(\varepsilon,\vec p)$ is evident Lorentz invariant obeying the
372: following normalization:
373: \begin{equation}
374: \int S_{DS}(\varepsilon,\vec p)d\varepsilon d^3p\;=\;\int
375: S_{DS}(p)d^4p\;=\; A
376: \end{equation}
377: (here $p_0=m+\varepsilon$, so $dp_0=d\varepsilon$).\\
378:
379: The calculations were performed for $^{12}C,\ ^{14}N,\ ^{27}Al,\
380: ^{40}Ca,\ ^{56}Fe$ and $^{63}Cu$. The reason is as follows. As
381: mentioned above the necessary information for the calculations is that
382: about the proton and neutron density distributions.
383: % That for
384: The former
385: is available for throughout the whole periodic system\cite{8b}, but it
386: is not the case for the latter: the neutron densities are available
387: only for doubly closed-shell nuclei $^{16}O,\ ^{40}Ca,\ ^{90}Zr$ and
388: $^{208}Pb$\cite{9b}. That's why we confined ourselves by nuclei with a
389: small neutron excess: the density distributions per nucleon are nearly
390: the same for protons and neutrons in these nuclei\cite{10b}.\\
391:
392: To calculate the eigen functions the Bonn B \cite{6b} and
393: OSBEP\cite{OSBEP} NN-potentials were used
394: \footnote{For the deuteron the Bonn B wave function was used in both
395: cases.}. In both cases the results are very close to each other.
396: The difference never exceeds 0.5\% for $x<0.6$ and is less then the
397: experimental error bars in the domain where the ratio (\ref{ratio})
398: $R_{th}>1$.
399:
400:
401: \section{Deep inelastic cross section\\ on nuclear target}
402: The DIS cross section is usually written in terms of the structure
403: function $F_2(x,Q^2)$, that is the cross section of electron-nucleon
404: interaction
405: \begin{equation}
406: \label{dis}
407: \frac{d\sigma}{dxdQ^2}\simeq\frac{4\pi\alpha^2}{xQ^4}
408: \left((1-y+\frac{y^2}2)F_2(x,Q^2)-\frac{y^2}2F_L(x,Q^2)\right)
409: \end{equation}
410: where we neglect the nucleon mass $m^2_N=m^2$ in comparison with the
411: total energy square $s=(k+p)^2 >> m^2$. Here: $k,q,p$ are the
412: 4-momenta of the incoming electron, heavy photon and the target nucleon
413: repectively. $Q^2=-q^2$, $x=Q^2/2(p\cdot q)$ and $y=(q\cdot p)/(k\cdot
414: p)$. $\alpha=1/137$ is the electromagnetic coupling.\\
415:
416: As a rule the data are taken at rather small $y$,
417: where the coefficient $y^2/2$ in front of the longitudinal part
418: ($F_L$) is small. Next, the ratio $R^L=F_L/F_2 \sim 0.2$ is not large.
419: Moreover, unlike the $F_2$, the function $R^L$ does not appear to depend
420: on atomic number $A$\cite{Arn}.
421:
422: Therefore the ratio of cross sections is given usually in terms of
423: the ratio of structure functions $F_2$.\\
424:
425: In order to compare our results with the data, where experimentalists
426: already accounted for the difference between proton and neutron,
427: we write the structure function of nucleus as
428: $$\frac 1A
429: F_{2A}(x,Q^2)=\frac 1A\left( ZF_{2pA}+NF_{2nA}\right)=$$
430: \begin{equation}
431: \label{iso}
432: =\frac{F_{2nA}(x,Q^2)+F_{2pA}(x,Q^2)}2 +\frac{N-Z}{2A}(F_{2nA}-F_{2pA})
433: \end{equation}
434: and select the isospin I=0 part of $F_2$ given by the first term of
435: (\ref{iso}). The ratio which we will discuss reads
436: \begin{equation}
437: \label{ratio}
438: R_{kin}(x,Q^2)=\frac{F_{2nA}(x,Q^2)+F_{2pA}(x,Q^2)}{F_{2D}(x,Q^2)}\; .
439: \end{equation}
440:
441: The structure function of the proton in nucleus
442: \begin{equation}
443: \label{f2p}
444: F_{2pA}(x,Q^2)=\frac 1Z\sum^{(p)}_\lambda
445: \nu_{p\lambda}F_{2p\lambda}(x,Q^2)\; ,
446: \end{equation}
447: where $\nu_{p\lambda}$ is the actual number of protons on the level
448: $\lambda$ ($\nu_{p\lambda}=2j+1$ for the completely ocupied
449: shell). Note that in the experimental data the variable $x$ was
450: calculated assuming the proton momentum $p_N$ equal to the momentum of
451: a free proton at rest, $p_N=(m_N,0,0,0)$. However to single out the
452: precise "kinematics" one must account for the change of the nucleon
453: structure function (parton distibutions) in medium caused by the change
454: of the Bjorken variable $x=Q^2/2(p\cdot q)$.
455: In other words calculating the momentum fraction $x'$ carried by the
456: quark we need to use the precise four
457: momentum of the nucleon in medium. That is
458: \begin{equation}
459: x'=\frac{Q^2}{2(pq)}\;=\;\frac{Q^2}{2(p_0q_0-\vec p\vec q)}\;=\;
460: \frac {mx}{m+\varepsilon_{\lambda}-\beta p t}\; ,
461: \label{xn}
462: \end{equation}
463: where $\beta=|\vec q|/q_0=(1+\frac{4m^2x^2}{Q^2})^{1/2}$
464: and the variable $t$ is the cosine of the angle between $\vec p$
465: and $\vec q$.\\
466:
467: Next we have to note that the structure function $F_2$, which
468: at the LO reads
469: \begin{equation}
470: F_2=\sum_f e^2_f(xq_f(x)+x\bar q_f(x))
471: \label{F2}
472: \end{equation}
473: ( $e_f$ is the electric charge of the quark of flavour $f$)\\
474: contains two factors: the quark(antiquark) distribution $q(x)$($\bar
475: q(x)$) and the kinematical factor $x$. The origin of this kinematical
476: factor is as follows. The covariant quantity is not the cross section
477: but discontinuity of the dimensionless interaction amplitute
478: $ImA\simeq s\sigma$. Going from the amplitude $A$ to the cross section
479: $\sigma\propto 1/Q^2$ we obtain the factor $x_A=Q^2/2(pq)$ which
480: corresponds to the true nucleus target and must be calculated as
481: $x_A=AQ^2/2m_A\nu$, where $m_A$ is the mass of nucleus and $\nu=q_0$ is
482: the photon energy in the nucleus rest frame. Note that in the
483: final expressions (\ref{f2lam},\ref{f2D}) we use the value of
484: $m_A$ calculated within the doorway formalism,
485: that is $m_A=\sum_\lambda^{(p)}\nu_{p\lambda}(m+\varepsilon_\lambda)
486: +\sum_\lambda^{(n)}\nu_{n\lambda}(m+\varepsilon_\lambda)$,
487: which is about 3\%
488: less than the true mass of a nucleus in the ground
489: state\footnote{For example, the 'doorway' mass of $^{40}Ca$ is
490: $m(doorway)=0.968m(ground\; state)$}.
491: This is equivalent to the prescription given in \cite{FS},
492: where the authors accounted for the relativistic flux factor and used
493: the baryon charge conservation to normalize the spectral functions
494: $S_p(\varepsilon,\vec p)$ and $S_n(\varepsilon,\vec p)$. \\
495:
496: Thus in (\ref{f2p}) we need to calculate the function
497:
498: $$F_{2p\lambda}(x,Q^2)=\frac 12
499: \int^{p_{\lambda}}_0f_\lambda(p)p^2dp \int^1_{-1}
500: \frac{x_A}{x'}
501: F_{2p}(x',Q^2) dt\; +\;$$
502: \begin{equation}
503: \label{f2lam}
504: +\;\frac 12\int_{p_{\lambda}}^\infty f_\lambda(p)p^2dp
505: \int^{p_\lambda/p}_{-1}
506: \frac{x_A}{x'}
507: F_{2p}(x',Q^2) dt\; .
508: \end{equation}
509:
510: Here: $p_\lambda=((1-x)m+\varepsilon_\lambda)/\beta$,
511: and $f_\lambda(p)$
512: %=\varphi^2_\lambda(k)+\chi^2_\lambda(k)$
513: was defined in sect.2.2
514: \footnote{Strictly speaking (\ref{f2lam})
515: is correct for a positive $p_\lambda$ only. When $x$ is
516: close to 1 and $p_\lambda$ becomes negative one has to keep only the
517: last term in (\ref{f2lam}) with the integration from $-p_\lambda$ up
518: to $\infty$. In this case the values of $t < 0$ and $x'<x$. So the
519: quantity $F_{2\lambda}$ has non zero value even at $x=1$. Note however
520: that for experimentally available $x$ values the quantities
521: $p_\lambda$ never become negative.}\\
522:
523:
524: Exactly the same formulae is used for the neutron in nucleus.\\
525: For the deuteron
526: $$F_{2D}(x,Q^2)=\frac 12
527: \int^{p_D}_0f_D(p)p^2dp
528: \int^1_{-1}
529: \frac{x_D}{x'_D}
530: (F_{2p}(x'_D,Q^2)+F_{2n}(x'_D,Q^2))dt\; +$$
531: \begin{equation}
532: \label{f2D}
533: +\; \frac 12\int_{p_D}^\infty f_D(p)p^2dp
534: \int^{p_D/p}_{-1}
535: \frac{x_D}{x'_D}
536: (F_{2p}(x'_D,Q^2)+F_{2n}(x'_D,Q^2))dt
537: \end{equation}
538: with
539: $$x'_D=\frac {mx}{m_D-\sqrt{p^2+m^2} - \beta pt} $$
540: %(1+\frac{4m^2x^2}{Q^2})^{1/2}
541: %pm t}$$
542: and
543: % $p_D=((1-x)m+\varepsilon_D-\sqrt{p^2+m^2}+m)/\beta$;
544: $p_D=(\beta(m_D-mx)-\sqrt{(m_D-mx)^2+(\beta^2-1)m^2}\ )/(\beta^2-1)$;
545: $f_D(p)$ is just the sum of the squared monopole and quadrupole
546: components of the deuteron wave function;
547: $m_D$
548: % and $\varepsilon_D <0$ are
549: is the deuteron mass.
550: % and binding
551: % energy correspondingly.
552: Note that denominator in the expression for
553: $x'_D$ corresponds to the kinematics where the spectator nucleon is on
554: mass-shell.\\
555:
556:
557: The $F_{2p}(x,Q^2)$ and $F_{2n}(x,Q^2)$ free nucleon structure
558: functions were calculated using the MRST2002 NLO parametrization
559: \cite{MRST} obtained from the global parton analysis.
560:
561: \section{Discussion}
562: The results of calculations are presented in Table 1-5 and
563: Fig.1. The predictions made using the Bonn-B and OSBEP potentials
564: are very close to each other. So we present the results for the case
565: of Bonn B potential only.\\
566: %@@@@ Valery, please, include some details
567: %of the computations, if needed (if you would like) @@@@@\\
568: \begin{table}[htb]
569: $$\begin{array}[t]{|c c| c c| c|}
570:
571: \hline
572:
573: ^{12}C & & NA-037 & NMC\cite{-3} & \\
574: \hline
575: x & Q^2 & R_{exp} & (\pm) & R_{kin} \\
576: \hline
577: .125 & 12.0 & 1.032 & ( .012) & 0.997 \\
578: .175 & 15.0 & 1.011 & ( .015) & 0.994 \\
579: .250 & 20.0 & 1.010 & ( .015) & 0.990 \\
580: .350 & 27.0 & 0.971 & ( .020) & 0.985 \\
581: .450 & 32.0 & 0.975 & ( .029) & 0.985 \\
582: .550 & 37.0 & 0.925 & ( .043) & 0.999 \\
583: .650 & 41.0 & 0.873 & ( .064) & 1.052 \\
584: \hline
585: \end{array}$$
586: \caption{The ratio of structure functions $F^A_2$ measured on carbon
587: %$A=C$
588: to that on deuteron. The values of $Q^2$ are given in GeV$^2$.}
589: \end{table}
590:
591: \begin{table}[htb]
592: $$\begin{array}[t]{|c c| c c| c|}
593:
594: \hline
595: ^{14}N & & NA-4 & BCDMS\cite{-5} & \\
596:
597: \hline
598: x & Q^2 & R_{exp} & (\pm) & R_{kin} \\
599: \hline
600: .100 & 32.0 & 1.018 & ( .039) & 0.997 \\
601: .140 & 40.0 & 1.018 & ( .031) & 0.995 \\
602: .180 & 49.0 & 1.002 & ( .024) & 0.993 \\
603: .225 & 56.0 & 1.035 & ( .025) & 0.990 \\
604: .275 & 56.0 & 1.024 & ( .027) & 0.988 \\
605: .350 & 67.0 & 0.983 & ( .025) & 0.985 \\
606: .450 & 77.0 & 0.941 & ( .031) & 0.985 \\
607: .550 & 84.0 & 0.891 & ( .047) & 0.999 \\
608: .650 & 96.0 & 0.826 & ( .075) & 1.053 \\
609:
610:
611: \hline
612: \end{array}$$
613: \caption{The ratio of structure functions $F^A_2$ measured on nitrogen
614: %$A=N$
615: to that on deuteron. The values of $Q^2$ are given in GeV$^2$.}
616: \end{table}
617:
618: \begin{table}[htb]
619: $$\begin{array}[t]{|c c| c c| c|}
620:
621: \hline
622:
623: ^{40}Ca & & NA-037 & NMC\cite{-4} & \\
624:
625: \hline
626: x & Q^2 & R_{exp} & (\pm) & R_{kin} \\
627: \hline
628: .113 & 4.3 & 0.994 & ( .010) & 0.998 \\
629: .138 & 5.1 & 1.007 & ( .012) & 0.996 \\
630: .175 & 6.2 & 1.001 & ( .011) & 0.994 \\
631: .225 & 7.7 & 1.015 & ( .014) & 0.990 \\
632: .275 & 9.1 & 0.998 & ( .018) & 0.986 \\
633: .350 & 11.0 & 0.996 & ( .019) & 0.981 \\
634: .450 & 14.0 & 1.024 & ( .031) & 0.978 \\
635: .600 & 17.0 & 0.955 & ( .038) & 1.005 \\
636:
637:
638: \hline
639: \end{array}$$
640: \caption{The ratio of structure functions $F^A_2$ measured on calcium
641: %$A=Ca$
642: to that on deuteron. The values of $Q^2$ are given in GeV$^2$.}
643: \end{table}
644:
645: \begin{table}[htb]
646: $$\begin{array}[t]{|c c| c c| c|}
647:
648: \hline
649:
650: ^{56}Fe & & NA-4 & BCDMS\cite{-6} & \\
651:
652: \hline
653: x & Q^2 & R_{exp} & (\pm) & R_{kin} \\
654: \hline
655: .100 & 22.0 & 1.057 & (.021) & 0.996 \\
656: .140 & 25.0 & 1.046 & (.020) & 0.994 \\
657: .180 & 29.0 & 1.050 & (.018) & 0.991 \\
658: .225 & 46.0 & 1.027 & (.019) & 0.988 \\
659: .275 & 49.0 & 1.000 & (.021) & 0.984 \\
660: .350 & 59.0 & 0.959 & (.020) & 0.979 \\
661: .450 & 72.0 & 0.923 & (.028) & 0.977 \\
662: .550 & 72.0 & 0.917 & (.040) & 0.991 \\
663: .650 & 72.0 & 0.813 & (.053) & 1.047 \\
664: \hline
665: \end{array}$$
666: \caption{The ratio of structure functions $F^A_2$ measured on iron
667: %$A=Fe$
668: to that on deuteron. The values of $Q^2$ are given in GeV$^2$.}
669: \end{table}
670:
671: \begin{table}[htb]
672: $$\begin{array}[t]{|c c| c c| c|}
673:
674: \hline
675:
676: ^{63}Cu & & NA-037 & NMC\cite{-7} & \\
677:
678: \hline
679: x & Q^2 & R_{exp} & (\pm) & R_{kin} \\
680: \hline
681:
682: .123 & 11.0 & 1.041 & ( .026) & 0.996 \\
683: .173 & 16.1 & 1.031 & ( .023) & 0.993 \\
684: .243 & 19.3 & 1.018 & ( .024) & 0.988 \\
685: .343 & 25.8 & 0.962 & ( .032) & 0.981 \\
686: .444 & 36.0 & 0.959 & ( .047) & 0.978 \\
687: .612 & 46.4 & 0.918 & ( .056) & 1.016 \\
688:
689:
690: \hline
691: \end{array}$$
692: \caption{The ratio of structure functions $F^A_2$ measured on copper
693: %$A=Cu$
694: to that on deuteron. The values of $Q^2$ are given in GeV$^2$.}
695: \end{table}
696:
697:
698: Recall that here we \underline{assume} the parton distributions
699: inside the nucleon in nucleus to be the same as that for the free
700: nucleon and evaluate the pure kinematical effect of the boundness and
701: the motion of nucleon in nuclear matter. Using the doorway states,
702: which are the correct eigen functions to describe the fast interaction
703: with one nucleon, we account for the full \underline{4-momemtum}
704: of the (target) nucleon and for the exitation of the ``residual''
705: nucleus $(A-1)$.
706:
707:
708: Thus the difference between the calculated value of $R_{kin}$ and the
709: data indicates the distortion of the parton wave function of a nucleon
710: placed in nuclear medium.\\
711:
712: As expected the account of the boundness and Fermi motion of nucleons
713: in nuclei diminishes the cross section in the $x=0.2\ -\ 0.63$ interval.
714: Indeed, due to the boundness (and the fact that about 24 - 27 MeV is
715: spent for the exitation of the residual $(A-1)$ nucleus) the mean value
716: of shifted argument $x'$ (\ref{xn}) is larger than the value of
717: $x$ on a free nucleon. On the other hand in this domain the free
718: nucleon structure function $F_2$ falls down with $x$. Therefore we get
719: $R_{kin}<1$.\\
720:
721: At a large $x$, close to 1, the details of angular integration (over
722: $t$ in (\ref{f2lam})) become important. For a negative $t$, due to a
723: Fermi motion, there is a region where $x'<x$ (see (\ref{xn})). Thanks
724: to the contribution coming from this region the value of $R_{kin}$
725: becomes larger than 1 for $x>0.65\ -\ 0.7$.\\
726:
727:
728:
729: Clearly, besides the Fermi motion there should be some dynamical
730: effects. At a large $x$ the growth of the ratio $R(x,Q^2)$ with $x$ is
731: usually atributed to a short range nucleon-nucleon
732: correlations\cite{FS2} or to a multiquark bags\cite{bag} (see for a
733: details the reviews \cite{Arn,FS2} and reference therein). However,
734: contrary to the conventional expectations, the theoretical value of
735: $R_{kin}$ resulting after account of the Fermi motion in the doorway
736: states is even \emph{larger} than the value $R_{exp}$ measured
737: experimentally\footnote{Since the same effect was observed both at
738: relatively low $Q^2$ in SLAC data and for a larger $Q^2$ at CERN this
739: can not be explained by the account of the mass correction.}.
740:
741: This
742: means that in nuclear medium the (one nucleon) parton distribution
743: becomes softer, that is the probability to find a parton with $x>0.45$
744: inside the in-medium-nucleon is less than that in a free nucleon. In
745: other words in medium the quark distribution is shifted towards a
746: lower $x$, leading to the decrease of quark density at $x>0.45$ and a
747: larger quark density at a lower $x\sim 0.1\ -\ 0.2$.\\
748:
749:
750: Next at small $x<0.2$ the partons from different
751: (neighbouring) nucleons start to overlap and to interact with each other.
752: Indeed, according to uncertainty principle
753: the characteristic size of localization is $\Delta r\sim 1/mx$ and for
754: $x<0.2$ the value of $\Delta r > 1$fm
755: becomes comparable with the nucleon-nucleon separation.
756: At a very low $x$ the partons screen each other and this shadowing
757: correction results in decreasing of $R(x,Q^2)$. Another way to describe
758: this effect is to say that two low-$x$ partons from two different
759: nucleons recombine into one parton. However the whole energy must be
760: conserved. This leads to the antishadowing (growth of the parton
761: density)\cite{NZ} (see the reviews\cite{Arn,FS2} for more details)
762: just in the region ($x\sim 0.1\ -\ 0.2$) of the begining of
763: recombination. On the other hand this antishadowing effect is expected
764: to reveal itself more in the gluon distributions than in the quark
765: structure function.
766:
767: Thus it is not surprising that
768: %for $x<0.2-0.3$ the ratio $R_{exp}$
769: %observed experimentally is systematically a bit larger than our values
770: %of $R_{th}$.\\
771: in the interval $0.2<x<0.45$ the ratio given by the pure kinematical
772: effects $R_{kin}$ (\ref{ratio})is close (within the error bars) to that
773: observed experimentally $R_{exp}$.\\
774:
775:
776: Note that, at large $x$, the Fermi motion is not negligible, even
777: for the deuteron. The ratio $R_{D,kin}=F_{2D}/(F_{2p}+F_{2n})$
778: is close to one for $x<0.65$, but it noticeably differs from one
779: for $x>0.75$, reaching values of $R_{D,kin}=$1.07 (1.42) at
780: $x=$0.75 (0.85); see Table 6.
781:
782: \begin{table}[htb]
783: $$\begin{array}[t]{|c|r r r r r r c|}
784: \hline
785: x & Q^2=\; 5 & 10 & 20 & 50 & 100 & 200 & GeV^2 \\
786: \hline
787: % x & 5 & 10 & 20 & 50 & 100 & 200 \\
788: % \hline
789:
790: .05 & R_{kin}=\; 1.000 & .999 & .999 & .999 & .998 & .998 & \\
791: .10 & R_{kin}=\;\;\; .999 & .999 & .998 & .998 & .997 & .997 &
792: \\
793:
794: .14 & R_{kin}=\;\;\; .998 & .998 & .997 & .997 & .996 & .996 &
795: \\
796:
797: .20 & R_{kin}=\;\;\; .996 & .996 & .995 & .995 & .994 & .994 &
798: \\
799:
800: .35 & R_{kin}=\;\;\; .990 & .990 & .989 & .989 & .989 & .989 & \\
801:
802: .45 & R_{kin}=\;\;\; .987 & .987 & .987 & .987 & .987 & .987 &
803: \\
804:
805: .55 & R_{kin}=\;\;\; .988 & .988 & .988 & .989 & .990 & .991 &
806: \\
807: .65 & R_{kin}=\; 1.005 & 1.004 & 1.004 & 1.006 & 1.008 & 1.011
808: & \\
809: .75 & R_{kin}=\; 1.080 & 1.073 & 1.071 & 1.075 & 1.080 & 1.085 &
810: \\
811: .85 & R_{kin}=\; 1.440 & 1.396 & 1.382 & 1.391 & 1.408 & 1.431 &
812: \\
813:
814:
815: %0.05 & R_{kin}=\; ? & ? & & & & & \\
816: %0.10 & R_{kin}=\; ? & & & & & & \\
817: %.... & & & & & & & \\
818: \hline
819:
820: \end{array}$$
821: \caption{The kinematical part $R_{kin}$ of the ratio
822: $F^d_2/[F^p_2+F^n_2]$ calculated using the Bonn-B potential}
823: \end{table}
824:
825:
826: An analysis performed by the MRST group
827: shows that if this effect is included then one obtains practically the
828: same partons, but the description of the high $x$ deuteron data is much
829: improved; with $\chi^2$ reduced by 20 for the 12 deuteron data points that are fitted at
830: $x=0.75$ \footnote{We thank R.S.Thorne and A.D.Martin for
831: discussions and for performing a new analysis using our Fermi motion in
832: the deuteron.}.\\
833:
834: \begin{figure}
835: \centerline{\vspace{0.2cm}\hspace{0.1cm}
836: \epsfig{file=SLAC-C.eps,width=7cm}
837: \epsfig{file=SLAC-AL.eps,width=7cm}
838: }
839: \vspace*{-0.5cm}
840: \centerline{\vspace{0.2cm}\hspace{0.1cm}
841: \epsfig{file=SLAC-CA.eps,width=7cm}
842: \epsfig{file=SLAC-FE.eps,width=7cm}
843: }
844: \caption{ Fig.1 The ratio of structure function $F^A_2$
845: measured on nucleus $A$ to that on deuteron. $Q^2=5$ GeV$^2$.
846: The data are taken from \cite{SLAC}. The empty bar
847: is the ratio $R_{kin}$ calculated using the Bonn-B potential.}
848: \end{figure}
849:
850: After the present work was completed we have read the recent paper of
851: A.Molochkov \cite{Mol} where a little bit another (but not quite
852: different from that used here) prescripton was proposed to account for
853: the boundness and momentum distribution of nucleons. A.Molochkov had
854: considered the ratio of the $^4$He to deuteron structure functions.
855: The shortness of his prescription is the assumption that both the
856: nucleon structure function $F_2$ and the momentum distribution of the
857: nucleons in nucleus $f^N(P_A,p)$ are regular (i.e. have no
858: singularities) with respect to $p_0$. Besides this some terms,
859: coming from the differentiation of the nucleus
860: $(A-1)$ propagator and the factor $1/(p_0+\sqrt{m^2+p^2})^2$
861: (corresponding to the antinucleon pole) in the nucleon propagator,
862: which are proportional to the binding (or nuclear exitation) energy,
863: were omitted in \cite{Mol}. We hope that our approach, based
864: on the 'doorway' formalism is more precise. Moreover, in terms of
865: the Molochkov's integral our result may be
866: obtained by closing the integration contour over $p_0$ in the upper
867: half-plane (on the pole corresponding to the residue $(A-1)$ nucleus)
868: instead of the lower one as it was done in\cite{Mol}.
869: %by Molochkov.
870:
871: %In terms of the Molochkov's
872: %integral our result may be obtained closing the contour of $p_0$
873: %integration on the pole corresponding to the residue (A-1) nucleus.
874: %We hope that our approach, based
875: %on the 'doorway' formalism is more precise.
876:
877: However we are planning to
878: compare both approaches in the forthcoming paper, using the doorway
879: eigen functions to describe the distributions of nucleons in a heavier
880: nuclei.
881:
882:
883:
884:
885: %\section{Conclusion}
886: %....????
887:
888:
889:
890:
891:
892:
893:
894: \begin{thebibliography}{99}
895: \bibitem{EMC} J.J.Aubert et al., Phys. Lett. B123 (1983) 275.
896: \bibitem{BR} A.Bodek and J.L.Ritchie, Phys. Rev. D23 (1981) 1070; D24
897: (1981) 1400.
898: \bibitem{Vagr} S.V.Akulinichev et al., Phys. Rev. Lett. 55 (1985) 2239.
899: \bibitem{FS} L.Frankfurt and M.Strikman, Phys. Lett. B183 (1987) 254.
900: \bibitem{Arn} M.Arneodo, Phys. Rept. 240 (1994) 301.
901: \bibitem{6} S.Frullani and J.Mougey, Adv. Nucl. Phys. 14 (1984) 1.
902: \bibitem{3} B.L. Birbrair and V.I. Ryazanov, Yad. Fiz. {\bf 63}, 1842
903: (2000).
904: \bibitem{BR2} B.L. Birbrair and V.I. Ryazanov, Yad. Fiz. {\bf64}, 471
905: (2001).
906: \bibitem{BR3} B.L. Birbrair and V.I. Ryazanov, Eur. Phys. J.
907: {\bf A15} (2002) 343.
908:
909: \bibitem{BLev} B.L.Birbrair et al., Phys. Lett. B166 (1986) 119.
910: %\bibitem{Vagr} Vagradov ...
911: %\bibitem{Bir1} indoor states.
912: %\bibitem{Born} Bonn potential
913: %\bibitem{OSBEP} OSBEP...
914: %\bibitem{Vor} ....exper. nucleon density
915: \bibitem{7} A.A. Abrikosov, L.P. Gor'kov, and I.E.~Dzyaloshinsky,
916: {\em"Methods of quantum field theory in statistical physics"}
917: Prentice-Hall, Englewood Cliffs, 1963.
918:
919:
920: \bibitem{8} A.B. Migdal, {\em"Theory of finite Fermi-systems and
921: applications to atomic nuclei"}, Nauka, M. 1983.
922: \bibitem{6b} R.Machleidt, K.Holinde and Ch.Elster, Phys. Rept. 149
923: (1987) 1.
924: \bibitem{8b} H. de Vries et al., At. Data and Nucl. Data Tables
925: 36 (1987) 495.
926: \bibitem{9b} G.D.Alkhazov et al., Nucl. Phys. A381 (1982) 430.
927: \bibitem{10b} G.D.Alkhazov, private communication.
928:
929: %\bibitem{Born} Born potential
930: \bibitem{OSBEP} L.J\" ade and H.V. von Geramb, Phys. Rev. C57
931: (1998) 496.
932: \bibitem{MRST} A.D.Martin, R.G.Roberts, W.J.Stirling and R.S.Thorne,
933: Eur. Phys. J. C28 (2003) 455.
934: %\bibitem{NZ} N.N.Nikolaev and V.I.Zakharov, Yad. Fiz. 21 (1975) 434;
935: %Sov.J.Nucl.Phys. 21 (1975) 227; Phys. Lett. B55 (1975) 397; Z.Phys. C49
936: %(1991) 607.
937:
938:
939: %[-1] Ashman at al. Phys. Lett. 202B (1988) 603
940: %[-2] Amaudruz at al. Zeit. Phys. C51 (1991) 387
941: \bibitem{-3} M.Arneodo at al. Nucl. Phys. B441 (1995) 12.
942: \bibitem{-5} G.Bari at al. Phys. Lett. 163B (1985) 282.
943: \bibitem{-4} P.Amadruz at al. Nucl. Phys. B441 (1995) 3.
944: \bibitem{-6} A.C.Benvenuti et al. Phys. Lett. 189B (1987) 483.
945: \bibitem{-7} J.Ashman at al. Zeit. Phys. C57 (1993) 211.
946:
947:
948: \bibitem{SLAC} R.G.Arnold et al. Phys. Rev. Lett. 52 (1984) 727.
949:
950: \bibitem{FS2} L.L.Frankfurt and M.I.Strikman, Phys. Rept. 160 (1988)
951: 235.
952:
953: \bibitem{bag} R.L.Jaffe Phys. Rev. Lett. 50 (1983) 228;\\
954: C.E.Carlson and T.J.Havens, Phys. Rev. Lett. 51 (1983) 261;\\
955: S.Date and A.Nakamura, Prog. Theor. Phys. 69 (1983) 565;\\
956: H.Faissner and B.B.Kim, Phys. Lett. B130 (1983) 321;\\
957: H.J.Pirner and J.P.Vary, Phys. Rev. Lett. 46 (1981) 1376,
958: Nucl. Phys. A358 (1981) 413c.
959: \bibitem{NZ} N.N.Nikolaev and V.I.Zakharov, Yad. Fiz. 21 (1975) 434;
960: Sov.J.Nucl.Phys. 21 (1975) 227; Phys. Lett. B55 (1975) 397; Z.Phys. C49
961: (1991) 607.
962:
963: %\bibitem{4} H. de Vries {\em et al.}, At. Dana and Nucl. Data Tables
964: %{\bf36}, 495 (1987).
965:
966: %\bibitem{5} G.D. Alkhazov {\em et al.}, Nucl. Phys. A{\bf381}, 430
967: %(1982).
968:
969: %\bibitem{6} S.S. Volkov et al., Yad. Fiz. {\bf52}, 1339 (1990);\\
970: %A.A. Vorobyov {\em et al.}, Yad. Fiz. {\bf58}, 1923 (1995).
971: \bibitem{Mol} A.Molockov, arXiv:nucl-th/0407077
972:
973: \end{thebibliography}
974:
975: \end{document}
976:
977:
978: