1: \documentstyle[12pt,epsfig]{article}
2: \topmargin -48pt
3: \def\baselinestretch{1.2}
4: \textwidth 160mm
5: \textheight 235mm
6: \oddsidemargin 2mm
7: \evensidemargin 2mm
8: \begin{document}
9:
10: \title{Perspectives of $e^+e^-$ production in
11: $pp$, $pd$ and $p Be$ reactions at SIS energies
12: \thanks{Work supported by GSI and BMBF} }
13: \author{E. L. Bratkovskaya, W. Cassing, and U. Mosel \\[2mm]
14: {\normalsize Institut f\"{u}r Theoretische Physik, Universit\"{a}t Giessen}\\
15: {\normalsize 35392 Giessen, Germany}}
16: \date{}
17: \maketitle
18:
19: \begin{abstract}
20: We study dilepton production from $pp$, $pd$ and $p Be$ collisions from
21: 1 -- 5 GeV including the $\pi^0$, $\eta$, $\omega$ and $\Delta$ Dalitz
22: decays, direct decays of vector mesons ($\rho$, $\omega$) as well as
23: subthreshold $\rho$ production via baryonic resonances (e.g.
24: $D_{13}(1520), P_{11}(1710)$). Our calculations compare rather well
25: with the $pp$ and $pd$ data from the DLS Collaboration, however,
26: overestimate slightly the 'old' $p Be$ data from that group.
27: Futhermore, detailed predictions for differential dilepton spectra at
28: SIS energies are made with a high mass resolution that can be
29: controlled experimentally by the HADES Collaboration in near future.
30: \end{abstract}
31:
32: \vspace{0.5cm}\noindent
33: PACS: \ {25.75.Dw, 13.30.Ce, 12.40.Yx}
34:
35: \vspace{0.5cm}\noindent
36: Keywords: particle and resonance production; leptonic and semileptonic
37: decays; hadron models
38:
39: %--------------------------------------------------------------------
40: \newpage
41: \section{Introduction}
42:
43: Electromagnetic probes such as dileptons provide the most clear signals from
44: the early phases of high-energy heavy-ion collisions because they
45: may leave the reaction volume essentially undistorted by final-state
46: interactions. Differential dilepton spectra from heavy-ion collisions
47: thus can provide information about the effective degrees of freedom at
48: high baryon density and temperature. Apart from that the in-medium
49: properties of hadrons (cf. Refs.
50: \cite{BrownRho,H&L92,Shakin94,Klingl96,Asakawa93,mosel91}), that are
51: closely related to the latter problem, can be explored as well.
52: According to QCD sum rules \cite{H&L92,Asakawa93,Leupold} as well as
53: QCD inspired effective Lagrangian models
54: \cite{BrownRho,Shakin94,Klingl96,Herrmann,asakawa,Chanfray,Rapp,Friman,RappNPA,Peters}
55: especially the vector mesons ($\rho$, $\omega$ and $\phi$)
56: significantly change their properties with nuclear density. It has
57: been argued, that the experimentally observed enhanced production of
58: soft lepton pairs above known sources in nucleus-nucleus collisions at
59: SPS energies \cite{CERES,HELIOS} might be due to the in-medium
60: modification of vector mesons \cite{Li,Li96,Cass95CH,Brat97,CBRW97}
61: rather than reflecting a new state of hadronic matter.
62:
63:
64: Dileptons from heavy-ion collisions have also been measured by the DLS
65: Collaboration \cite{DLSold,DLSnew} at BEVALAC energies, where a
66: different temperature and density regime is probed. The first
67: generation of DLS data \cite{DLSold}, based on a limited data set, were
68: consistent with the results from early transport model calculations
69: \cite{Xiong90,Wolf90,Gudima,BCMas96} including the conventional
70: dileptons sources as $pn$ bremsstrahlung, $\pi^0$, $\eta$, $\omega$ and
71: $\Delta$ Dalitz decay, direct decay of vector mesons and pion-pion
72: annihilation, without incorporating any medium effects. A more recent
73: DLS measurement \cite{DLSnew} including the full data set and an
74: improved analysis, however, shows an increase by about a factor of 5-7
75: in the cross section in comparison to Ref. \cite{DLSold} and the early
76: theoretical predictions.
77:
78: This enhancement of the $e^+e^-$ yield in nucleus-nucleus collisions
79: provides some theoretical 'puzzle'. In Ref. \cite{BCRW97} the in-medium
80: $\rho$ spectral functions from Refs.~\cite{RappNPA,Peters} have been
81: implemented in the HSD transport approach for $\rho$ mesons produced in
82: baryon-baryon, pion-baryon collisions as well as from $\pi\pi$
83: annihilation, and a factor of 2-3 enhancement has been obtained
84: compared to the case of a free $\rho$-spectral function. In
85: Ref.~\cite{Ernst} dropping vector meson masses and $\omega$ meson
86: broadening were incorporated in the UrQMD transport model, which also
87: showed an enhancement of the dilepton yield, however, the authors could
88: not describe the new DLS data \cite{DLSnew} as well. Another attempt
89: to solve the DLS 'puzzle' has been performed in Ref.~\cite{BrKo98}
90: where the dropping hadron mass scenario was considered together with
91: the subthreshold $\rho$ production in $\pi N$ scattering via the
92: baryonic resonance $N(1520)$ whose importance was pointed out in
93: Refs.~\cite{Peters,Stony98}. It was found that the enhancement of the
94: dilepton spectra due to low mass $\rho$'s from the $N(1520)$ was not
95: sufficient to match the DLS data. Thus, all in-medium scenarios that
96: successfully have explained the dilepton enhancement at SPS energies
97: failed to describe the new DLS dilepton data \cite{DLSnew} from
98: heavy-ion collisions ($^{12}C + ^{12}C$ and $^{40}Ca + ^{40}Ca$).
99:
100: In 1998 the DLS Collaboration has, furthermore, published dilepton
101: data from elementary $pp$ and $pd$ collisions at 1-5~GeV \cite{DLSpp}
102: which provides the possibility for an independent check of the
103: elementary dilepton channels that enter as 'input' in transport
104: calculations for heavy-ion reactions. Such an analysis has been
105: carried out in Refs.~\cite{Ernst,Brat99} and it was shown that the
106: dilepton invariant mass spectra from $pp$ reactions in the energy
107: range from 1 -- 5 GeV could be rather well described within the
108: standard sources of hadronic Dalitz and direct decays when including
109: also the 'subthreshold' production of $\rho$-mesons by baryonic
110: resonances \footnote{For a detailed description of the individual
111: channels we refer the reader to Ref. \cite{Brat99}.}. It has been,
112: furthermore, argued that additional channels (like dilepton decays of
113: scalar mesons ($a_0, f_0$) or four-body final states decays like
114: $\eta^\prime\to \pi^+\pi^-e^+e^-$, $(\rho^0,\omega)\to
115: \pi^+\pi^-e^+e^-$, $(\rho^0,\omega)\to \eta \pi^0 e^+e^-$) might
116: contribute to the elementary yield in $pp$ and $pd$ reactions
117: \cite{Tuebingen}. Thus it is indispensible, that also these elementary
118: reactions have to be measured again with high resolution in order to
119: understand more accurately the $e^+e^-$ radiation mechanisms.
120:
121:
122: In this article we perform a detailed study of dilepton production from
123: $pp$, $pd$ and $p Be$ collisions from 1 -- 5 GeV including all channels
124: as described in Ref. \cite{Brat99}, i.e. the subthreshold $\rho$
125: production via baryonic resonances (e.g. $D_{13}(1520), P_{11}(1710)$)
126: as well as $\pi^0$, $\eta$, $\omega$ and $\Delta$ Dalitz decays and the
127: direct decay of vector mesons ($\rho$, $\omega$, $\phi$). We
128: investigate the role of isospin degrees of freedom with respect to
129: dilepton production in $pd$ reactions by employing exact
130: energy-momentum conservation for the deuteron target. The $p Be$
131: reactions are calculated within the resonance transport model
132: \cite{Effe99gam} that includes the same elementary production
133: amplitudes.
134:
135: The paper is organized as follows: In Section~2 we briefly present the
136: underlying resonance model that enters the coupled-channel BUU
137: transport approach. In Section~3 we provide a comparison of our
138: results with the corresponding DLS data including the DLS acceptance
139: and resolution functions. In Section 4 we, furthermore, give detailed
140: predictions for these reactions with a high mass resolution in view of
141: upcoming experiments with the HADES detector at GSI Darmstadt. We close
142: with a summary and discussion of open problems in Section 5.
143:
144: %----------------------------------------------------------------------
145: \section{Description of the model}
146:
147: \subsection{Resonance approach}
148:
149: We perform our analysis of dilepton production from $pp$, $pd$
150: and $pBe$ collisions within the resonance approach of
151: Refs. \cite{Effe99gam,EffePhD}.
152: This model is based on the resonance concept of nucleon-nucleon and
153: meson-nucleon interactions at low invariant energy $\sqrt{s} \ $
154: \cite{TeisZP97} by adopting all resonance parameters from the Manley
155: analysis \cite{Manley};
156: all states with at least 2 stars in Ref.~\cite{Manley}
157: are taken into account:
158: $P_{33}$(1232), $P_{11}$(1440), $D_{13}$(1520), $S_{11}$(1535),
159: $P_{33}$(1600), $S_{31}$(1620), $S_{11}$(1650), $D_{15}$(1675),
160: $F_{15}$(1680), $P_{13}$(1879), $S_{31}$(1900), $F_{35}$(1905),
161: $P_{31}$(1910), $D_{35}$(1930), $F_{37}$(1950), $F_{17}$(1990),
162: $G_{17}$(2190), $D_{35}$(2350). These resonances couple to the following
163: channels: $N \pi$, $N \eta$, $N \omega$, $\Lambda K$, $\Delta(1232)
164: \pi$, $N \rho$, $N \sigma$, $N(1440) \pi$, $\Delta(1232) \rho$
165: with respect to the production and decay.
166:
167: It has been shown, that the resonance model provides a good description
168: of the experimental data on one- and two-pion production in
169: nucleon-nucleon collisions at low energy \cite{TeisZP97}. However, with
170: increasing energy the resonance contributions underestimate the data;
171: the missing yield is then treated as a backgraund term to the resonance
172: amplitude. This background term 'mimics' $t$-channel particle
173: production mechanism as well as other non-resonance contributions (e.g.,
174: direct $NN\to NN\pi$, without creating an intermediate resonance).
175:
176: With increasing energy, furthermore, the multiparticle production
177: becomes more and more important.
178: The high energy collisions -- above $\sqrt{s}$ = 2.6~GeV for
179: baryon-baryon collisions and $\sqrt{s}$ = 2.2~GeV for meson-baryon
180: collisions -- are described by the LUND string fragmentation model
181: FRITIOF \cite{FRITIOF}. This aspect is similar to that used in the HSD
182: approach \cite{Ehehalt,Brat97,Geiss,CB99PR} and the UrQMD model
183: \cite{Bass}.
184:
185: This combined resonance-string approach allows to calculate
186: particle production in baryon-baryon and meson-baryon
187: collisions from low to high energies.
188: The collisional dynamics for proton-nucleus reactions, furthermore, is
189: described by the coupled-channel BUU transport approach
190: \cite{Effe99gam,EffePhD} that is based on the same elementary cross
191: sections.
192:
193:
194: \subsection{Dilepton production}
195:
196: The dilepton production within the resonance model can be schematically
197: presented in the following way:
198: \begin{eqnarray}
199: BB &\to&R X \label{chBBR} \\
200: mB &\to&R X \label{chmBR} \\
201: && R \to e^+e^- X, \label{chRd} \\
202: && R \to m X, \ m\to e^+e^- X, \label{chRMd} \\
203: && R \to R^\prime X, \ R^\prime \to e^+e^- X. \label{chRprd}
204: \end{eqnarray}
205: In words: in a first step a resonance $R$ might be produced
206: in baryon-baryon ($BB$) or meson-baryon ($mB$) collisions -- (\ref{chBBR}),
207: (\ref{chmBR}). Then this resonance can couple to dileptons
208: directly -- (\ref{chRd}) (e.g., Dalitz decay of the $\Delta$ resonance:
209: $\Delta \to e^+e^-N$) or decays to a meson $m$ (+ baryon) -- (\ref{chRMd})
210: which produces dileptons via direct decays ($\rho, \omega$)
211: or Dalitz decays ($\pi^0, \eta, \omega$).
212: The resonance $R$ might also decay into another resonance $R^\prime$ --
213: (\ref{chRprd}) which later produces dileptons via Dalitz decay or again
214: via meson decays (e.g., $D_{35}(1930)\to\Delta\rho,\ \Delta\to
215: e^+e^-N, \ \rho\to e^+e^-$). Note, that in our combined model the
216: final particles -- which couple to dileptons -- can be produced also
217: via non-resonant mechanisms, i.e. 'background' at low and intermediate
218: energies and string decay at high energies.
219:
220: The electromagnetic part of all conventional dilepton sources --
221: $\pi^0, \eta, \omega$ and $\Delta$ Dalitz decay, direct decay of vector
222: mesons $\rho, \omega$ -- are treated in the same way as described in
223: detail in Sec. 3 of Ref.~\cite{Brat99} where we have calculated the
224: dilepton spectra from $pp$ collisions in comparison to the DLS data.
225: The only difference is that in our present calculation we use the
226: spectral function of vector mesons ($\rho, \omega$) in the same form
227: as for all other resonances as used in Manley analysis \cite{Manley}:
228: \begin{equation}
229: f(M) = N_V \ {2\over \pi} \ {M^2 \Gamma^V_{tot}
230: \over (M^2-m_V^2)^2 + (M {\Gamma_{tot}^V})^2}.
231: \label{fmdist}\end{equation}
232: In (\ref{fmdist}) $N_V$ guarantees normalization to unity, i.e. $\int
233: f(M) dM =1$, while $\Gamma_{tot}^V$ is the total vector meson width.
234:
235: For $pd$ and $pBe$ collisions we also take into account $pn$
236: bremstrahlung using the soft-photon approximation \cite{Gale87}, where
237: the radiation from internal meson lines is neglected and the strong
238: interaction vertex is assumed to be on-shell. In this case the strong
239: interaction part and the electromagnetic part separate; however, the
240: cross section for dileptons has to be corrected
241: \cite{GaleKap89,Schaefer89} by reducing the phase-space for the
242: colliding particles in their final-state.
243: We discard $pp$ bremsstrahlung from the nucleon pole since
244: the microscopic OBE calculations \cite{Schaefer,BrTit,Fred}
245: have shown that at 1.0 GeV the $pp$ bremsstrahlung is smaller than
246: the $\Delta$ Dalitz decay contribution by a factor of 2-3.
247: At this energy also the interference between nucleon and $\Delta$-pole
248: terms is negligible. At high energies, where this interference
249: becomes important \cite{Schaefer}, however, the overall contribution from
250: these channels is negligible (see below). We, furthermore, discard
251: $e^+e^-$ production from the scalar mesons $f_0, a_0$ as advocated in
252: Ref.~\cite{Tuebingen} since our studies in Ref.~\cite{Brat_Kon}
253: have indicated that the multiplicity of scalar mesons in $pp$ reactions
254: is very small in the energy regime of interest.
255:
256: The basic underlying assumption for dilepton production, that enters in
257: our calculation, is that the amplitude can be factorized into a meson
258: or/and resonance production and a dilepton decay part.
259: This assumption provides not only a significant simplification, which
260: is necessary for applications in transport calculations
261: \cite{Wolf90,CB99PR,Effe99gam}, but also allows to take into account the
262: 'inelasticity' from many-particle production channels which become
263: dominant at high energy.
264:
265: In Fig.~\ref{Fig1SIS} we illustrate our factorization assumption for
266: the process $NN\to RN \to \rho NN \to e^+e^- NN$ with an intermediate
267: resonance $R$ coupled to the $\rho$ meson, e.g. $D_{13}(1520)$,
268: which is responsible for $\rho$ production below the threshold.
269: As shown in Fig.~\ref{Fig1SIS} we cut the diagram twice such that we
270: have a resonance production part ($NN\to RN$), a resonance decay to the
271: $\rho$ meson ($R\to \rho N$) and direct dilepton decay of the $\rho$ meson
272: ($\rho\to e^+e^-$). This assumption with respect to subthreshold $\rho$
273: production via the $D_{13}(1520)$ is different to the ansatz used in
274: the previous dilepton analysis \cite{BrKo98,Brat99}. In Ref. \cite{Brat99}
275: we have factorized the amplitude $NN\to RN \to \rho NN \to e^+e^- NN$ into
276: a $D_{13}(1520)$ production part ($NN\to RN$) and Dalitz decay part
277: $R\to \rho N \to e^+e^-N$ (cf. Sect. 3.1 of Ref. \cite{Brat99}).
278: In the latter approach the $\rho$ meson is considered to be off-shell, i.e.
279: a virtual particle, and can have a mass below $2 m_\pi$ which leads to
280: the Dalitz behavior of the dilepton spectra (cf. Fig. 3 of Ref.
281: \cite{Brat99}), whereas in the present approach we assume that the $\rho$
282: meson has a free spectral function with masses only above $2m_\pi$.
283: The latter assumption leads to the 'direct $\rho$ decay' profile for
284: dileptons from subthreshold $\rho$'s and is consistent with our
285: resonance prescription used also in the transport calculations
286: \cite{Effe99gam}, where we propagate such $\rho$ mesons explicitly.
287:
288: To compare with the experimental data of the DLS Collaboration one has
289: to use the appropriate experimental filter, which is a function of the
290: dilepton invariant mass $M$, transverse momentum $p_T$ and rapidity
291: $y_{lab}$ in the laboratory frame -- $F(M,p_T,y_{lab})$. For that
292: purpose we explicitly simulate all dileptons by Monte-Carlo
293: and apply the filter $F(M,p_T,y_{lab})$.
294:
295:
296: %----------------------------------------------------------------------
297: \section{$e^+e^-$ results in comparison to DLS data}
298:
299: \subsection{Dileptons from $pp$ collisions}
300:
301: In Fig.~\ref{Fig2SIS} we present the calculated dilepton invariant mass
302: spectra $d\sigma/dM$ for $pp$ collisions from 1.0 -- 4.9 GeV including
303: the final mass resolution and filter $F(M,p_T,y_{lab})$ from the DLS
304: Collaboration (version 4.1) in comparison to the DLS data \cite{DLSpp}.
305: The thin lines indicate the individual contributions from the different
306: production channels; {\it i.e.}~ starting from low $M$:
307: Dalitz decay $\pi^0 \to \gamma e^+ e^-$ (short dashed line),
308: $\eta \to \gamma e^+ e^-$ (dotted line),
309: $\Delta \to N e^+ e^-$ (dashed line),
310: $\omega \to \pi^0 e^+ e^-$ (dot-dashed line),
311: for $M \approx $ 0.7 GeV: $\omega \to e^+e^-$ (dot-dashed line),
312: $\rho^0 \to e^+e^-$ (short dashed line).
313: The full solid line represents the sum of all sources considered here.
314:
315:
316: Whereas at 1.04 GeV the dileptons stem practically all from $\pi^0$ and
317: $\Delta$ Dalitz decays, $\rho$ decay becomes more important with
318: increasing energy and at high invariant mass.
319: Note, that in the present analysis we do not separate the production
320: mechanism for $\rho$ mesons as in Ref.~\cite{Brat99}, i.e. the
321: $\rho$-meson contribution might come from the decay of $D_{13}(1520)$
322: and $P_{11}(1710)$ or non-resonance $\rho$ production. Thus, the short
323: dashed line (denoted as $\rho^0 \to e^+e^-$) presents the sum of all
324: $\rho$ contributions. However, the resonance
325: mechanism dominates for $\rho$ production at low energies, especially
326: below the $pp\to \rho pp$ threshold $\sqrt s < 2m_N+m_\rho$, where
327: $\sqrt s$ is the invariant collision energy (see Sec. 2 in
328: Ref.~\cite{Brat99}).
329:
330: At 2.1 and 4.9 GeV the dilepton yield is dominated by the $\eta$ Dalitz
331: decay and direct decays of $\rho$ and $\omega$ mesons. The $\rho$
332: spectrum is enhanced towards low $M$ due to the limited phase space and
333: strong mass dependence $(M^{-3})$ of the dilepton decay width
334: \cite{Brat99} in line with the vector dominance model.
335: Note, that at 2.1 GeV we use the resonance production mechanism in
336: our model, whereas at 4.9 GeV most of the particles are produced by
337: string decays. Within the strict resonance model which gives
338: only the exclusive $pp\to pp\eta$ cross section the $\eta$ yield from
339: $pp$ collisions is underestimated by about a factor of 2
340: (dot-dot-dashed line) at 2.1 GeV. Since at this energy other
341: channels such as $NN\eta\pi, NN\eta\pi\pi$ are open, we have
342: estimated the cross section for these channels by the string model and
343: added it to the total $\eta$ yield. This gives the final $\eta$
344: contribution (dotted line) for 2.1 GeV.
345:
346:
347: \subsection{Dileptons from $pd$ collisions}
348:
349: The experimental measurement of dilepton spectra from proton and
350: deuteron targets provides a unique possibility to get information about
351: the isospin dependence of the $NN$ interaction which is interesting in
352: itself and, furthermore, important with respect to heavy-ion reactions.
353: The $pp$ and $pd$ dilepton data from the DLS Collaboration \cite{DLSpp}
354: are the first experimental step in this direction.
355:
356: Since the deuteron is a bound system of proton and neutron, the Fermi
357: motion of the nucleons plays an important role for all interactions. For
358: example, Fermi motion leads to a shift of the threshold for
359: particle production in $pd$ interactions to lower $\sqrt{s}$, i.e. the
360: channels closed by kinematics in $pp$ collisions might be
361: open in $pd$. Thus, the $pd$ reactions can not be directly approximated
362: by the simple sum of $pp$ and $pn$. On the other hand, one can't
363: treat the deuteron within the usual mean-field approach used in transport
364: calculations for more heavy nuclei.
365:
366: In our analysis we thus take into account the Fermi motion of nucleons using
367: the momentum distribution of the deuteron calculated with
368: the Paris potential \cite{Paris}.
369: We initialize a deuteron in momentum space (using a Monte Carlo method)
370: according to the momentum distribution $f(p)$:
371: \begin{eqnarray}
372: f(p) = {|\Psi(p)|^2\over |\Psi(0)|^2},
373: \label{deut}\end{eqnarray}
374: where $p$ is the momentum of a nucleon in the deuteron center-of-mass,
375: while $\Psi(p)$ is the deuteron wave function.
376: To fulfill the binding energy constraint we assume
377: \begin{eqnarray}
378: E_{spec}=\sqrt{p^2+m_N^2}, \ \ E_{part}=E_d-E_{spec}.
379: \label{disper}\end{eqnarray}
380: The nucleon, which does not participate in the reaction, i.e. the spectator,
381: has a free dispersion relation for the energy-momentum $E_{spec}$,
382: whereas $E_{part}$ is the energy of the nucleon participating
383: in the reaction (participant).
384: In (\ref{disper}) $E_d = 2m_N+\varepsilon$ is the total deuteron energy,
385: where $\varepsilon=-2.2$ MeV is the binding energy of the deuteron.
386:
387: The invariant collision-energy distribution $dN/d\sqrt s$ for $pd$
388: reactions calculated with the dispersion relation (\ref{disper}) is
389: shown by the solid histogram in Fig.~\ref{Fig3SIS}. For comparison
390: (dashed histogram) we present the $dN/d\sqrt s$ distribution in the
391: physical region ($2m_N \le \sqrt s \le \sqrt{s_{max}}$) calculated
392: with $f(p)$ (\ref{deut}) in the quasi-free scattering limit, i.e. using
393: the free dispersion relation for both nucleons in the deuteron:
394: $E_{spec}=E_{part}=\sqrt{p^2+m_N^2}$.
395: The lower limit of $\sqrt s$ is given by the free cross section that
396: only exists for $\sqrt \ge 2m_N$.
397: As seen from Fig.~\ref{Fig3SIS}, both prescriptions are the same in the
398: peak value, however, the deuteron spectral distribution (\ref{disper})
399: leads to an enhancement of the $\sqrt s$ distribution at small $\sqrt
400: s$ and a reduction at high $\sqrt s$. This implies a substantial
401: lowering of particle production at high $\sqrt s$ with respect to the
402: quasi-free limit, which has been often used in the literature.
403:
404: Within the 'binding' prescription (\ref{disper}) we have calculated the
405: dilepton invariant mass spectra $d\sigma/dM$ for $pd$ collisions from
406: 1.0 -- 4.9 GeV including the final mass resolution and
407: filter $F(M,p_T,y_{lab})$ from the DLS Collaboration (version 4.1).
408: The results are show in Fig.~\ref{Fig4SIS} in comparison to the DLS data
409: \cite{DLSpp}, where the assignment of the lines is the same as in
410: Fig.~\ref{Fig2SIS}. Additionally, the dot-dot-dashed lines indicate the
411: $pn$ bremsstrahlung contributions.
412:
413: Similar to Fig.~\ref{Fig2SIS}, at 1.04 GeV the $\pi^0$ and $\Delta$
414: dalitz decays are dominant, however, due to the larger mass of the
415: deuteron, the $\eta$ channel is already open contrary to the $pp$
416: reaction and the subthreshold $\rho$ branch is getting more important
417: at high invariant mass $M$. Also $pn$ bremsstrahlung contributes at low
418: energies, however, with increasing bombarding energy it becomes less
419: important.
420:
421: Starting from 1.61 GeV the dominant contribution at low $M$ is again
422: the $\eta$ Dalitz decay, whereas at high $M$ the dileptons stem
423: basically from direct $\rho, \omega$ decays.
424: The $\eta$ contribution here is calculated by assuming
425: $\sigma_{pn\to pn\eta}=6\sigma_{pp\to pp\eta}$ for
426: $2.425\ {\rm GeV} \le \sqrt s \le 2.517\ {\rm GeV}$ in line with
427: the experimental data from Ref. \cite{Pinot},
428: $\sigma_{pn\to pn\eta}=2.5\sigma_{pp\to pp\eta}$ for
429: $2.517\ {\rm GeV} \le \sqrt s \le 2.65\ {\rm GeV}$ and
430: $\sigma_{pn\to pn\eta}\simeq \sigma_{pp\to pp\eta}$ for
431: $\sqrt s \ge 2.65\ {\rm GeV}$ where the multiparticle production
432: (above 2.65 GeV) is evaluated within the LUND string model. Note,
433: that at high $\sqrt s$ the isospin asymmetry $\sigma_{pn}/\sigma_{pp}$
434: should be washed out by the isospin of pions that are created together
435: with the $\eta$. Though this 'recipe' seems to work in comparison to
436: the DLS data, we point out that further experimental data with good
437: mass resolution will be necessery to clarify the situation for the
438: isospin couplings.
439:
440: In order to demonstrate the importance of isospin degrees-of-freedom for
441: dilepton production we present in Fig.~\ref{Fig5SIS} the invariant
442: mass spectra $d\sigma/dM$ for $pp$ (short dashed line), $pn$ (long dashed
443: line), $pp+pn$ (dot-dashed line) and $pd$ (solid line) collisions at
444: 1.61 GeV in comparison to the DLS data \cite{DLSpp}.
445: At these energies the $pn$ channel gives the main
446: contribution to the $pd$ dilepton yield. Furthermore, the influence of
447: the Fermi motion in the deuteron is clearly seen at high invariant mass $M$
448: by comparing the dot-dashed ($pp+pn$) and solid ($pd$) lines.
449:
450: In Fig. \ref{Fig6SIS} we present the excitation function, i.e. the
451: integrated dilepton cross section for masses above 0.15 GeV,
452: as a function of the available energy $Q=\sqrt{s}-2m_p$ for $pp$
453: (dashed line) and $pd$ (solid line) in comparison to the DLS data
454: \cite{DLSpp} (open circles for $pp$ and full circles for $pd$).
455: The calculated excitation functions for $pp$ and $pd$ increase
456: with the energy and stay within the experimental errorbars.
457:
458:
459: \subsection{Dileptons from $pBe$ collisions}
460:
461: Since $Be$ is one of the lightest nuclei, in-medium effects play
462: a minor role because most hadron decays occur already in vacuum.
463: So one can consider $pBe$ collisions as an additional
464: and independent check for the elementary reactions and isospin degrees
465: of freedom. However, the collisional dynamics has to be taken into
466: account properly. For that purpose we use the coupled-channel BUU
467: transport approach as discussed in Sec. 2.
468:
469: Dileptons from $pBe$ have been measured by the DLS Collaboration in the
470: end of the 80's \cite{DLSold}. In Fig.~\ref{Fig7SIS} we show the
471: calculated dilepton invariant mass spectra $d\sigma/dM$ for $pBe$
472: collisions at 1.0 (upper part), 2.1 (middle part) and 4.9 GeV (lower
473: part) including the final mass resolution and filter
474: $F(M,p_T,y_{lab})$ from the DLS Collaboration (version 1.6) in
475: comparison to the data \cite{DLSold}. The assignment of the lines is
476: the same as in Figs.~\ref{Fig2SIS} and \ref{Fig4SIS}. With the filter
477: 1.6 we do not get enough suppression for the dilepton yield at $M\le
478: 0.15$ GeV which basically stems from $\pi^0$ Dalitz decay. Similar to
479: $pd$ reactions the dominant channels at 1.0 GeV are the $\Delta$ Dalitz
480: decay and $pn$ bremsstrahlung. At 2.1 and 4.9 GeV the $\eta$ Dalitz
481: decay becomes important at low $M$ and the direct decays of vector
482: mesons ($\rho, \omega$) for $M\sim 0.8$ GeV. Within the present model
483: we slightly overestimate the data at 4.9 GeV, however, we discard any
484: discussion about the physical reason since the 'old' DLS filter is not
485: sufficiently understood. For example, the 'old' DLS data for $pBe$ at
486: 4.9 GeV show a similar dilepton yield as for $pd$, whereas according to
487: our calculations it should be higher by a factor of $\sim 2$ since more
488: nucleons are involved in the $pBe$ reaction.
489:
490: In this respect we have performed detailed calculations for $pp$, $pd$
491: and $pBe$ reactions without implementing any filter, however, including
492: a high (10 MeV) mass resolution for $d\sigma/dM$. For the results,
493: that can be controlled by the HADES Collaboration in near future, we
494: refer to the following Section.
495:
496: %----------------------------------------------------------------------
497: \section{Predictions for HADES}
498:
499: In this Section we present our predictions for the dilepton invariant
500: mass spectra, transverse momentum ($p_T$) distributions and rapidity
501: ($y$) distributions for $pp$, $pd$ and $pBe$ collisons at 1.5 GeV and
502: 4.0 GeV, which will be measured with the HADES spectrometer at GSI.
503:
504: In Fig.~\ref{Fig8SIS} the calculated dilepton invariant mass spectra
505: $d\sigma/dM$ for $pp$ (upper part), $pd$ (middle part) and $pBe$
506: collisions (lower part) at 1.5 GeV (left panel) and 4.0 GeV (right
507: panel) are shown including a $\Delta M=10$~MeV mass resolution. The
508: assignment of the lines is the same as in Figs.~\ref{Fig2SIS} and
509: \ref{Fig4SIS}. The dominant contribution at low $M$ ($> m_{\pi^0}$) is
510: the $\eta$ Dalitz decay, however, for $M > 0.4$ GeV the dileptons stem
511: basically all from direct vector meson decays ($\rho$ and $\omega$).
512: Within a 10 MeV mass resolution the peak from the direct $\omega$ decay
513: (which was completely smeared out within the DLS mass resolution) is
514: seen. The $\omega$ peak is getting more pronounced at high energy (cf.
515: the right panel of Fig.~\ref{Fig8SIS}).
516:
517: In Fig.~\ref{Fig9SIS} we present the transverse momentum distribution
518: $d\sigma/dp_T /(2\pi p_T)$ for $pp$ (upper part), $pd$ (middle part)
519: and $pBe$ collisions (lower part) at 1.5 GeV (left panel) and 4.0 GeV
520: (right panel). The assignment of the lines again is the same as in
521: Figs.~\ref{Fig2SIS} and \ref{Fig4SIS}.
522: Here we did not imply any cut in the invariant mass $M$. For all
523: systems and energies considered the $\pi^0$ Dalitz decay is much
524: higher than all other contributions, which is in line with the result of
525: Fig.~\ref{Fig8SIS} because the pion production cross section
526: is much larger than that for the other mesons. The shape of the $\pi^0,
527: \eta, \Delta, \omega$ Dalitz decays is defined by the kinematics of the
528: Dalitz decay and the 'flat' part at low $p_T$ depends on the mass of
529: the decaying resonance.
530:
531: The laboratory rapidity distributions $d\sigma/dy$ for $pp$ (upper
532: part), $pd$ (middle part) and $pBe$ collisions (lower part) at 1.5 GeV
533: (left panel) and 4.0 GeV (right panel) are shown in
534: Fig.~\ref{Fig10SIS}, where the assignment of the lines is the same as
535: before. The situation is similar to Fig.~\ref{Fig9SIS}, -- the $\pi^0$
536: Dalitz decay is the only visible contribution when imposing no cuts
537: on the invariant mass.
538:
539: In order to make the contribution from other channels visible one has
540: to suppress the 'background' $\pi^0$ yield. In Fig.~\ref{Fig11SIS} we
541: display the transverse momentum spectra for $pp$ (upper part), $pd$
542: (middle part) and $pBe$ collisions (lower part) at 1.5 GeV (left panel)
543: and 4.0 GeV (right panel) using a cut in invariant mass as 0.4~GeV$\le
544: M \le 0.7$~GeV. Fig.~\ref{Fig12SIS} shows the laboratory rapidity
545: distributions for the same systems and kinematical conditions. With
546: this cut on intermediate dilepton masses one can select the region
547: where the $\pi^0$ Dalitz decay is totally removed and the $\eta$ Dalitz
548: decay is suppressed substantially (cf. Fig.~\ref{Fig8SIS}); in this way
549: the $\rho$ branch in the transverse momentum and rapidity spectra
550: becomes seen (cf. Figs.~\ref{Fig10SIS}, \ref{Fig11SIS}). Thus, a
551: measurement of the transverse momentum and rapidity spectra with a
552: corresponging cut in $M$ will provide additional and independent
553: information about the $\rho$ contribution to the dilepton spectra.
554:
555: Finally, in Fig.~\ref{Fig13SIS} we present the double differential
556: dilepton spectra $d\sigma/dMdp_T$ as a function of the invariant mass
557: $M$ and transverse momentum $p_T$ for $pd$ collisions at 4.0 GeV.
558: At fixed $p_T$ one can recognize the shape of the invariant mass
559: spectra (cf. Fig.~\ref{Fig8SIS}) with a strong $\pi^0$ Dalitz decay
560: branch at low $M$ as well as the contributions from $\eta$ Dalitz decay
561: and the vector meson ($\rho, \omega$) decays. At fixed $M$ the shape looks
562: similar to the one in Fig.~\ref{Fig9SIS}. At low $M$ the
563: exponential decrease stems from $\pi^0$ Dalitz decay, then the
564: spectra become flatter due to the contributions from $\eta$ Dalitz
565: ($M\le 0.4$ GeV) and direct $\rho$ decays. At $M\sim 0.78$ GeV the
566: peak from the direct decay of $\omega$ mesons is visible. Thus, such
567: type of 3-dimensional experimental information (or even 4-dimensional
568: including rapidity) allows to select the contributions from different
569: channels.
570:
571:
572: %----------------------------------------------------------------------
573: \section{Summary}
574:
575: Within the framework of the combined resonance-string approach we have
576: performed a detailed study of dilepton production from $pp$, $pd$ and
577: $pBe$ collisions at 1.0 -- 4.9 GeV. In this approach the particle
578: production proceeds through resonances at low energies and by string
579: formation and decay at high energies. The collisional dynamics is
580: described by the coupled-channel BUU transport approach. As dilepton
581: sources we have considered $\pi^0$, $\eta$, $\omega$ and $\Delta$
582: Dalitz decays, $pn$ bremsstrahlung and direct decays of vector mesons
583: ($\rho$, $\omega$).
584:
585: We have compared our calculated results to the experimental data of the
586: DLS Collaboration which provide constraints on the different individual
587: production channels. It has been found that the DLS data for the $pp$
588: and $pd$ collisions can be reasonably well described in our model
589: whereas for $pBe$ systems (especially at 4.9 GeV) our calculations give
590: a slightly higher dilepton yield. We have demonstrated the importance
591: to measure the dileptons from $pp$ and $pd$ (or even $pBe$) collisions
592: simultaneously since such data can provide constraints on the isospin
593: dependence of $pp$ and $pn$ interactions, which is important for an
594: understanding of heavy-ion data.
595:
596: Dilepton spectra from elementary reactions will be measured in future by
597: the HADES detector at GSI Darmshtadt with high mass resolution and good
598: accuracy. In this respect we have made predictions for the dilepton
599: invariant mass spectra, transverse momentum and rapidity distributions
600: for $pp$, $pd$ and $pBe$ at 1.5 and 4.0 GeV. We showed that proper
601: cuts in invariant mass for transverse momentum and rapidity spectra
602: allow to select different dilepton sources and to study, for example,
603: the $\rho$ meson channel in more detail. We have indicated,
604: furthermore, that it might be very useful to provide experimentally
605: multi-dimensional information, e.g. double differential dilepton
606: spectra $d\sigma/dMdp_T$, in order to investigate the individual
607: contributions.
608:
609:
610: \section*{Acknowledgements}
611: The authors are grateful for valuable discussions with J. Friese,
612: M. Kagarlis and A. Sibirtsev.
613:
614: %----------------------------------------------------------------------
615: \begin{thebibliography}{99}
616: \bibitem{BrownRho}
617: G.E. Brown and M. Rho, Phys. Rev. Lett. 66 (1991) 2720.
618: \bibitem{H&L92}
619: T. Hatsuda and S. Lee, Phys. Rev. C 46 (1992) R34.
620: \bibitem{Shakin94}
621: C.M. Shakin and W.-D. Sun, Phys. Rev. C 49 (1994) 1185.
622: \bibitem{Klingl96}
623: F. Klingl and W. Weise, Nucl. Phys. A 606 (1996) 329;
624: F. Klingl, N. Kaiser and W. Weise, Nucl. Phys. A 624 (1997) 527.
625: \bibitem{Asakawa93}
626: M. Asakawa and C.M. Ko, Phys. Rev. C 48 (1993) R526.
627: \bibitem{mosel91}
628: U. Mosel, Ann. Rev. Nucl. Part. Sci. 41 (1991) 29.
629: \bibitem{Leupold}
630: S. Leupold, W. Peters and U. Mosel, Nucl. Phys. A 628 (1998) 311.
631: \bibitem{Herrmann}
632: M. Herrmann, B. Friman, and W. N\"orenberg, Nucl. Phys. A 560
633: (1993) 411.
634: \bibitem{asakawa}
635: M. Asakawa, C. M. Ko, P. L\'evai, and X. J. Qiu, Phys. Rev.
636: C 46 (1992) R1159.
637: \bibitem{Chanfray}
638: G. Chanfray and P. Schuck, Nucl. Phys. A 545 (1992) 271c.
639: \bibitem{Rapp}
640: R. Rapp, G. Chanfray, and J. Wambach, Phys. Rev. Lett. 76
641: (1996) 368.
642: \bibitem{Friman}
643: B. Friman and H. J. Pirner, Nucl. Phys. A 617 (1997) 496.
644: \bibitem{RappNPA}
645: R. Rapp, G. Chanfray and J. Wambach, Nucl. Phys. A 617
646: (1997) 472.
647: \bibitem{Peters}
648: W. Peters, M. Post, H. Lenske, S. Leupold, and U. Mosel,
649: Nucl. Phys. A 632 (1998) 109.
650: \bibitem{CERES}
651: G. Agakichiev et al., Phys. Rev. Lett. 75 (1995) 1272;
652: Th. Ullrich et al., Nucl. Phys. A 610 (1996) 317c;
653: A. Drees, Nucl. Phys. A 610 (1996) 536c.
654: \bibitem{HELIOS}
655: M. A. Mazzoni, Nucl. Phys. A 566 (1994) 95c;
656: M. Masera, Nucl. Phys. A 590 (1995) 93c;
657: T. {\AA}kesson et al., Z. Phys. C 68 (1995) 47.
658: \bibitem{Li}
659: G. Q. Li, C. M. Ko, and G. E. Brown, Phys. Rev. Lett. 75
660: (1995) 4007.
661: \bibitem{Li96}
662: C. M. Ko, G. Q. Li, G. E. Brown, and H. Sorge,
663: Nucl. Phys. A 610 (1996) 342c.
664: \bibitem{Cass95CH}
665: W. Cassing, W. Ehehalt, and C. M. Ko, Phys. Lett. B 363 (1995) 35;
666: W. Cassing, W. Ehehalt, and I. Kralik, Phys. Lett. B 377 (1996) 5.
667: \bibitem{Brat97}
668: E. L. Bratkovskaya and W. Cassing, Nucl. Phys. A 619 (1997) 413.
669: \bibitem{CBRW97}
670: W. Cassing, E. L. Bratkovskaya, R. Rapp, and J. Wambach,
671: Phys. Rev. C 57 (1998) 916.
672: \bibitem{DLSold}
673: G. Roche et al., Phys. Rev. Lett. 61 (1988) 1069;
674: C. Naudet et al., Phys. Rev. Lett. 62 (1989) 2652;
675: G. Roche et al., Phys. Lett. B 226 (1989) 228.
676: \bibitem{DLSnew}
677: R.J. Porter et al., Phys. Rev. Lett. 79 (1997) 1229.
678: \bibitem{Xiong90}
679: L. Xiong, Z. G. Wu, C. M. Ko, and J. Q. Wu,
680: Nucl. Phys. A 512 (1990) 772.
681: \bibitem{Wolf90}
682: Gy. Wolf, G. Batko, W. Cassing, U. Mosel, K. Niita, and M. Sch\"afer,
683: Nucl. Phys. A 517 (1990) 615;
684: Gy. Wolf, W. Cassing and U. Mosel, Nucl. Phys. A 552 (1993) 549.
685: \bibitem{Gudima}
686: K.K. Gudima, A.I. Titov and V.D. Toneev,
687: Sov. Jour. of Nucl. Phys. 55 (1992) 1715.
688: \bibitem{BCMas96}
689: E. L. Bratkovskaya, W. Cassing and U. Mosel,
690: Phys. Lett. B 376 (1996) 12.
691: \bibitem{BCRW97}
692: E. L. Bratkovskaya, W. Cassing, R. Rapp, and J. Wambach,
693: Nucl. Phys. A 634 (1998) 168.
694: \bibitem{Ernst}
695: C. Ernst, S. A. Bass, M. Belkacem, H. St\"ocker, and W. Greiner,
696: Phys. Rev. C 58 (1998) 447.
697: \bibitem{BrKo98}
698: E.L. Bratkovskaya and C.M. Ko, Phys. Lett. B 445 (1999) 265.
699: \bibitem{Stony98}
700: G. E. Brown, G. Q. Li, R. Rapp, M. Rho, and J. Wambach,
701: Acta Phys. Polon. B 29 (1998) 2309.
702: \bibitem{DLSpp}
703: W.K. Wilson et al., Phys. Rev. C 57 (1998) 1865.
704: \bibitem{Brat99}
705: E. L. Bratkovskaya, W. Cassing, M. Effenberger, U. Mosel,
706: Nucl. Phys. A 653 (1999) 301.
707: \bibitem{Tuebingen}
708: A. Faessler, C. Fuchs and M.I. Krivoruchenko,
709: Phys. Rev. C 61 (2000) 035206.
710: \bibitem{Effe99gam}
711: M. Effenberger, E. Bratkovskaya and U. Mosel,
712: Phys. Rev. C 60 (1999) 044614.
713: \bibitem{EffePhD}
714: M. Effenberger, Ph.D. Thesis, Univ. of Giessen, 1999;
715: http://theorie.physik.uni-giessen.de/ftp.html.
716: \bibitem{TeisZP97}
717: S. Teis, W. Cassing, M. Effenberger, A. Hombach, U. Mosel,
718: and Gy. Wolf, Z. Phys. A 356 (1997) 421;
719: Z. Phys. A 359 (1997) 297.
720: \bibitem{Manley}
721: D.M. Manley and E.M. Saleski, Phys. Rev. D 45 (1992) 4002.
722: \bibitem{FRITIOF}
723: B. Anderson, G. Gustafson and Hong Pi, Z. Phys. C 57 (1993) 485.
724: \bibitem{Ehehalt}
725: W. Ehehalt and W. Cassing, Nucl. Phys. A 602 (1996) 449.
726: \bibitem{Geiss}
727: J. Geiss, W. Cassing, and C. Greiner, Nucl. Phys. A 644 (1998) 107.
728: \bibitem{CB99PR}
729: W. Cassing and E. L. Bratkovskaya, Phys. Rep. 308 (1999) 65.
730: \bibitem{Bass}
731: S. A. Bass et al., Prog. Part. Nucl. Phys. 42 (1998) 279;
732: J. Phys. G 25 (1999) 1859.
733: \bibitem{Gale87}
734: C. Gale and J. Kapusta, Phys. Rev. C 35 (1987) 2107;
735: C 40 (1987) 2397.
736: \bibitem{GaleKap89}
737: C. Gale and J. Kapusta, Nucl. Phys. A 495 (1989) 423c.
738: \bibitem{Schaefer89}
739: M. Sch\"afer, T. S. Biro, W. Cassing, and U. Mosel,
740: Phys. Lett. B 221 (1989) 1.
741: \bibitem{Schaefer}
742: M. Sch\"afer, H.C. D\"onges, A. Engel and U. Mosel,
743: Nucl. Phys. A 575 (1994) 429.
744: \bibitem{BrTit}
745: A.I. Titov, B. K\"ampfer and E.L. Bratkovskaya,
746: Phys. Rev. C 51 (1995) 227.
747: \bibitem{Fred}
748: F. de Jong and U. Mosel, Phys. Lett. B 379 (1996) 45;
749: B 392 (1997) 273.
750: \bibitem{Brat_Kon}
751: E. L. Bratkovskaya, W. Cassing, L. A. Kondratyuk, and A. Sibirtsev,
752: Eur. Phys. J. A 4 (1999) 165.
753: \bibitem{Paris}
754: M. Lacombe et al., Phys. Lett. B 101 (1981) 139.
755: \bibitem{Pinot}
756: E. Chiavassa et al., Phys. Lett. B 322 (1994) 270; B 337 (1994) 192;
757: H. Cal\'en et al., Phys. Rev. C 58 (1998) 2667.
758: \end{thebibliography}
759:
760: %\end{document}
761: %---------------------------------------------------------------------
762: \newpage
763:
764: \begin{figure}[t]
765: \phantom{a}\vspace*{2cm}
766: \centerline{\psfig{figure=fig1sis.eps,width=16cm}}
767: \phantom{a}\vspace*{2cm}
768: \caption{Factorization of the diagram for the process
769: $NN\to RN \to \rho NN \to e^+e^- NN$ with an intermediate
770: baryon resonance $R$.}
771: \label{Fig1SIS}
772: \end{figure}
773:
774:
775: \phantom{a}\newpage
776: \begin{figure}[h]
777: \centerline{\psfig{figure=fig2sis.eps,width=13cm}}
778: \caption{The calculated dilepton invariant mass spectra $d\sigma/dM$
779: for $pp$ collisions from 1.0 -- 4.9 GeV in comparison to the DLS data
780: \protect\cite{DLSpp}. The thin lines indicate the individual
781: contributions from the different production channels; {\it i.e.}~
782: starting from low $M$: Dalitz decay $\pi^0 \to \gamma e^+ e^-$ (short
783: dashed line), $\eta \to \gamma e^+ e^-$ (dotted line), $\Delta \to N
784: e^+ e^-$ (dashed line), $\omega \to \pi^0 e^+ e^-$ (dot-dashed
785: line); for $M \approx $ 0.7 GeV: $\omega \to e^+e^-$
786: (dot-dashed line), $\rho^0 \to e^+e^-$ (short dashed line). The
787: full solid line represents the sum of all sources considered here.
788: At 2.09 GeV the dot-dot-dashed line indicates the $\eta$ contribution from
789: the resonance decays only while the dotted line corresponds to
790: the inclusive $\eta$ production (see text).}
791: \label{Fig2SIS}
792: \end{figure}
793:
794: \begin{figure}[t]
795: \phantom{a}\vspace*{-2cm}
796: \centerline{\psfig{figure=fig3sis.eps,width=14cm}}
797: \caption{The distribution in the invariant collision energy $\sqrt{s}$
798: ($dN/d\sqrt s$) in arbitrary units for the $pd$ reaction at 1.04 GeV
799: calculated with the dispersion relation (\protect\ref{disper})
800: (solid histogram) and within the quasi-free scattering limit
801: (dashed histogram).}
802: \label{Fig3SIS}
803: \end{figure}
804:
805: \begin{figure}[t]
806: \phantom{a}\vspace*{-2cm}
807: \centerline{\psfig{figure=fig4sis.eps,width=13cm}}
808: \caption{The dilepton invariant mass spectra $d\sigma/dM$ for
809: $pd$ collisions from 1.0 -- 4.9 GeV in comparison to the DLS data
810: \protect\cite{DLSpp}. The assignment of the lines is the same as in
811: Fig.~\protect\ref{Fig2SIS}. Additionally, the dot-dot-dashed lines
812: indicate the $pn$ bremstrahlung contributions.}
813: \label{Fig4SIS}
814: \end{figure}
815:
816: \begin{figure}[t]
817: \centerline{\psfig{figure=fig5sis.eps,width=14cm}}
818: \caption{The dilepton invariant mass spectra $d\sigma/dM$ for
819: $pp$ (short dashed line), $pn$ (long dashed line), $pp+pn$ (dot-dashed
820: line) and $pd$ (solid line) collisions at 1.61 GeV in comparison
821: to the DLS data \protect\cite{DLSpp}. The difference between the
822: dot-dashed line and the solid line stems from the Fermi motion of
823: the nucleons in the deuteron.}
824: \label{Fig5SIS}
825: \end{figure}
826:
827: \begin{figure}[t]
828: \centerline{\psfig{figure=fig6sis.eps,width=14cm}}
829: \caption{The excitation function for dileptons from $pp$
830: (dashed line) and $pd$ (solid line) in comparison to the DLS data
831: \cite{DLSpp} (open circles for $pp$ and full circles for $pd$ reactions).}
832: \label{Fig6SIS}
833: \end{figure}
834:
835: \begin{figure}[t]
836: \phantom{a}\vspace*{-2cm}
837: \centerline{\psfig{figure=fig7sis.eps,width=10cm}}
838: \caption{The dilepton invariant mass spectra $d\sigma/dM$ for $pBe$
839: collisions at 1.0 GeV (upper part), 2.1 GeV (middle part) and 4.9 GeV
840: (lower part) in comparison to the DLS data \cite{DLSold}.
841: The assignment of the individual lines is the same as in
842: Figs.~\protect\ref{Fig2SIS}, \protect\ref{Fig4SIS}.}
843: \label{Fig7SIS}
844: \end{figure}
845:
846: \begin{figure}[t]
847: \phantom{a}\vspace*{-2cm}
848: \centerline{\psfig{figure=fig8sis.eps,width=13cm}}
849: \caption{The dilepton invariant mass spectra $d\sigma/dM$ for
850: $pp$ (upper part), $pd$ (middle part) and $pBe$ collisions (lower part)
851: at 1.5 GeV (left panel) and 4.0 GeV (right panel) including a 10 MeV
852: mass resolution. The assignment of the individual lines is the same as
853: in Figs.~\protect\ref{Fig2SIS}, \protect\ref{Fig4SIS}.}
854: \label{Fig8SIS}
855: \end{figure}
856:
857: \begin{figure}[t]
858: \phantom{a}\vspace*{-2cm}
859: \centerline{\psfig{figure=fig9sis.eps,width=13cm}}
860: \caption{The transverse momentum distribution $d\sigma/dp_T /(2\pi p_T)$
861: for $pp$ (upper part), $pd$ (middle part) and $pBe$ collisions (lower
862: part) at 1.5 GeV (left panel) and 4.0 GeV (right panel).
863: The assignment of the individual lines is the same as
864: in Figs.~\protect\ref{Fig2SIS}, \protect\ref{Fig4SIS}.}
865: \label{Fig9SIS}
866: \end{figure}
867:
868: \begin{figure}[t]
869: \phantom{a}\vspace*{-2cm}
870: \centerline{\psfig{figure=fig10sis.eps,width=13cm}}
871: \caption{The rapidity distributions $d\sigma/dy$ for $pp$ (upper part),
872: $pd$ (middle part) and $pBe$ collisions (lower part) at 1.5 GeV
873: (left panel) and 4.0 GeV (right panel).
874: The assignment of the individual lines is the same as in
875: Figs.~\protect\ref{Fig2SIS}, \protect\ref{Fig4SIS}.}
876: \label{Fig10SIS}
877: \end{figure}
878:
879: \begin{figure}[t]
880: \phantom{a}\vspace*{-2cm}
881: \centerline{\psfig{figure=fig11sis.eps,width=13cm}}
882: \caption{The transverse momentum distribution $d\sigma/dp_T /(2\pi p_T)$
883: for $pp$ (upper part), $pd$ (middle part) and $pBe$ collisions
884: (lower part) at 1.5 GeV (left panel) and 4.0 GeV (right panel)
885: implying a cut in invariant mass of $0.4 \le M \le 0.7$ GeV.
886: The assignment of the individual lines is the same as in
887: Figs.~\protect\ref{Fig2SIS}, \protect\ref{Fig4SIS}.}
888: \label{Fig11SIS}
889: \end{figure}
890:
891: \begin{figure}[t]
892: \phantom{a}\vspace*{-2cm}
893: \centerline{\psfig{figure=fig12sis.eps,width=13cm}}
894: \caption{The laboratory rapidity distributions $d\sigma/dy$ for $pp$
895: (upper part), $pd$ (middle part) and $pBe$ collisions (lower part)
896: at 1.5 GeV (left panel) and 4.0 GeV (right panel)
897: implying a cut in invariant mass of $0.4 \le M \le 0.7$ GeV.
898: The assignment of the individual lines is the same as in
899: Figs.~\protect\ref{Fig2SIS}, \protect\ref{Fig4SIS}.}
900: \label{Fig12SIS}
901: \end{figure}
902:
903: \begin{figure}[t]
904: \phantom{a}\vspace*{-2cm}
905: \centerline{\psfig{figure=fig13sis.eps,width=13cm}}
906: \caption{The calculated double differential dilepton spectra
907: $d\sigma/dMdp_T$ as a function of invariant mass $M$ and transverse
908: momentum $p_T$ for $pd$ collisions at 4.0 GeV.}
909: \label{Fig13SIS}
910: \end{figure}
911:
912: \end{document}
913:
914:
915: #!/bin/csh -f
916: # this uuencoded Z-compressed .tar file created by csh script uufiles
917: # for more information, see e.g. http://xxx.lanl.gov/faq/uufaq.html
918: # if you are on a unix machine this file will unpack itself:
919: # strip off any mail header and call resulting file, e.g., figures.uu
920: # (uudecode ignores these header lines and starts at begin line below)
921: # then say csh figures.uu
922: # or explicitly execute the commands (generally more secure):
923: # uudecode figures.uu ; uncompress figures.tar.Z ;
924: # tar -xvf figures.tar
925: # on some non-unix (e.g. VAX/VMS), first use an editor to change the
926: # filename in "begin" line below to figures.tar_Z , then execute
927: # uudecode figures.uu
928: # compress -d figures.tar_Z
929: # tar -xvf figures.tar
930: #
931: uudecode $0
932: chmod 644 figures.tar.Z
933: zcat figures.tar.Z | tar -xvf -
934: rm $0 figures.tar.Z
935: exit
936:
937: