1: %%%%%%%%%%%%%%%%%%%%%%% file LW.tex %%%%%%%%%%%%%%%%%%%%%%%%%
2: %
3: \begin{filecontents}{leer.eps}
4: gsave
5: 72 31 moveto
6: 72 342 lineto
7: 601 342 lineto
8: 601 31 lineto
9: 72 31 lineto
10: showpage
11: grestore
12: \end{filecontents}
13: %
14: \documentclass[epj]{svjour}
15: \usepackage{graphics}
16: \usepackage{amssymb}
17: \usepackage{amsmath}
18: \usepackage[dvips]{epsfig}
19: \newcommand{\fmn}[2]{\mbox{${\textstyle \frac{#1}{#2}}$}}
20: \newcommand{\dd}{\mbox{\rm d}}
21: \def\theequation{\thesection.\arabic{equation}}
22: \newcommand{\half}{\mbox{${\textstyle \frac{1}{2}}$}}
23:
24: \authorrunning{Fran\c{c}ois Lehar, Colin Wilkin}
25:
26: %
27: \begin{document}
28: %
29: \title{Nucleon charge exchange on the deuteron: A critical review}
30: \author{Fran\c{c}ois Lehar\inst{1}$^,$\inst{2}
31: \and
32: Colin Wilkin\inst{3}}
33: \institute{SPP IRFU, CEA Saclay, F-91190 Gif-sur-Yvette Cedex, France
34: \and IEAP CTU, Horsk\'a 3a/22, Cz-12800 Prague 2, Czech Republic,
35: \email{lehar@mail.utef.cvut.cz}
36: \and Physics and Astronomy Department, UCL, Gower Street, London WC1E 6BT, United Kingdom,
37: \email{cw@hep.ucl.ac.uk}}
38: %
39: \date{Received: \today}
40: %
41: \abstract{The existing experimental data on the $d(n,p)nn$ and
42: $d(p,n)pp$ cross sections in the forward direction are reviewed in
43: terms of the Dean sum rule. It is shown that the measurement of
44: the ratio of the charge exchange on the deuteron to that on the
45: proton might, if taken together with other experimental data,
46: allow a direct construction of the $np\to np$ scattering amplitude
47: in the backward direction with few ambiguities.
48: %
49: \PACS{
50: {13.75.Cs}{Nucleon nucleon interactions} \and
51: {25.40.Kv}{Charge-exchange reactions} \and
52: {25.10.+s}{Nuclear reactions involving few-nucleon systems}
53: } % end of PACS codes
54: } %end of abstract
55: %
56: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
57: %
58: \maketitle
59: %
60: \section{Introduction}
61: \label{intro} \setcounter{equation}{0}
62:
63: The nucleon-nucleon interaction is fundamental to the whole of
64: nuclear physics and hence to the composition of matter as we know
65: it. Apart from its intrinsic importance, it is also a necessary
66: ingredient in the description of meson production and other
67: intermediate energy processes.
68:
69: In the case of proton-proton scattering, the data set of
70: differential and total cross sections and the various single and
71: multi-spin observables is very extensive and this has allowed the
72: construction of reliable isospin $I=1$ phase shifts up to at least
73: 2\,GeV~\cite{SAID}. The situation is far less developed for the
74: isoscalar $I=0$ case, where the corresponding phase shift analysis
75: is only available up to 1.3\,GeV and even then there are
76: significant ambiguities at the higher energies~\cite{SAID}.
77:
78: More good data on neutron-proton scattering are clearly needed,
79: possibly with the aim of directly reconstructing the isosinglet
80: amplitudes. This is particulary promising in the forward
81: direction~\cite{Ball1} but the conditions are almost as favourable
82: for backward $pn\to np$ scattering (often loosely called the $np$
83: charge-exchange region) and it is this which we want to consider
84: in some detail in this paper.
85:
86: To avoid some of the problems associated with the quality of
87: neutron beams and/or the detection of neutron, the deuteron is
88: often used successfully as a substitute target or beam. For
89: example, it has been shown that the spin correlation and transfer
90: parameters in $pp$ quasi-elastic scattering in the 1.1 to 2.4~GeV
91: range are very close to those measured in free $pp$
92: collisions~\cite{Ball2} and the Saclay group find exactly the same
93: reassurance for $pn$ quasi-elastic scattering~\cite{Lesquen}.
94:
95: One particular valuable tool that can be used to study the
96: backward amplitudes is the comparison of the quasi-free $(p,n)$ or
97: $(n,p)$ reaction on the deuteron to the free backward elastic
98: scattering on a nucleon target. It was emphasised over 50 years
99: ago that the reaction on the deuteron can act, in suitable
100: kinematic regions, as a spin filter that selects the
101: spin-dependent contribution to the $np$ elastic cross
102: section~\cite{Pomeranchuk}. This sensitivity arises from the Pauli
103: principle, which blocks any spin-1 component in the low energy
104: $\{nn\}$ or $\{pp\}$ system. To avoid the explicit introduction of
105: the dynamics of the low energy $NN$ system, Dean~\cite{Dean}
106: derived a sum rule for the ratio $R_{np}$ of the differential
107: cross section for charge exchange on the deuteron to that on the
108: nucleon. Although this is given as a function of the momentum
109: transfer $q$, it simplifies for collinear dynamics to the extent
110: that there is then no dependence on the deuteron structure. More
111: importantly, the sum rule converges very fast as a function of the
112: excitation energy for small $q$, due to the strength of the low
113: energy $^{1\!}S_0$ $NN$ interaction. It has therefore been used in
114: the analysis of the wealth of $R_{np}$ data, which now extend up
115: to 2\,GeV~\cite{Sharov08}. It is the aim of the present paper to
116: show how such data, combined with other measurements, might
117: contribute to an analysis of the elastic neutron-proton scattering
118: amplitudes in the backward direction.
119:
120: In section~\ref{Amps} we summarise the amplitudes and some
121: observables that are relevant for backward elastic $np$
122: scattering. Of particular importance in this context is the fact
123: that the conventional $NN$ amplitudes~\footnote{See
124: Ref.~\cite{BYS78} for a very comprehensive discussion of different
125: amplitude bases.} are not the most suitable ones when analysing
126: charge exchange on the deuteron, where it is necessary to take
127: into account of the interchange between the final neutron and
128: proton \textit{ab initio}. The impulse approximation dynamics and
129: the form of the Dean sum rule in the forward direction are
130: described in section~\ref{deuteron}, where some of the underlying
131: assumptions are clarified. It is shown there that the sum rule
132: saturates very quickly, which make it such a useful tool.
133:
134: Section~\ref{data} gives an extensive compilation of the values of
135: $R_{np}(0)$ derived from the $nd \to p\{nn\}$, $pd\to n\{pp\}$,
136: and $dp\to \{pp\} n$ reactions and it is shown there that the
137: total error bars are the smallest in the $(n,p)$ case provided
138: that a good quality neutron beam is available. On the basis of the
139: existing phase shift analysis~\cite{SAID}, impulse approximation
140: predictions of $R_{np}(0)$ can be made up to a beam energy of
141: 1.3\,GeV and the agreement with experimental data is very
142: reasonable down to at least 300\,MeV. However, it is important to
143: reiterate that values of $R_{np}(0)$ are now available up to
144: 2\,GeV. The prospects for a $np \to np$ elastic amplitude
145: reconstruction in the backward direction are discussed in the
146: conclusions of section~\ref{summary}, where it is shown that, with
147: extra information available through the use of polarised
148: deuterons, this is now becoming feasible.
149:
150: %
151: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
152: %
153: \section{Neutron-proton amplitudes and observables}
154: \label{Amps} \setcounter{equation}{0}
155:
156: The nucleon-nucleon formalism, including the four-index notation
157: and definition of all \textit{pure} experiments, is discussed in
158: full detail in Ref.~\cite{BYS78}. In this work the matrix
159: describing
160: elastic neutron-proton scattering is written in the form%
161: \begin{eqnarray}%
162: \nonumber M(\vec{k}_f,\vec{k}_i) &=& \half\big[(a + b) + (a -
163: b)(\vec{\sigma}_n\cdot\hat{\vec{n}})(\vec{\sigma}_p\cdot\hat{\vec{n}})\\
164: \nonumber
165: &&\hspace{-2cm}+(c+d)(\vec{\sigma}_n\cdot\hat{\vec{m}})
166: (\vec{\sigma}_p\cdot\hat{\vec{m}})
167: +(c-d)(\vec{\sigma}_n\cdot\hat{\vec{\ell}})(\vec{\sigma}_p\cdot
168: \hat{\vec{\ell}})\\ &&+e(\vec{\sigma}_n + \vec{\sigma}_p)\cdot\hat{\vec{n}}\big]\,,%
169: \label{BYS}
170: \end{eqnarray}%
171: where $a$, $b$, $c$, $d$ and $e$ are complex invariant amplitudes,
172: which are functions of energy and scattering angle $\theta$. The
173: $2\times 2$ Pauli matrices $\vec{\sigma}_n$ and $\vec{\sigma}_p$
174: act in the spaces of the proton and neutron spins, respectively.
175:
176: In terms of the c.m.\ momenta in the initial and final states,
177: $\vec{k}_i$ and $\vec{k}_f$, an orthonormal basis system is
178: defined through%
179: \begin{equation}\label{basis1}%
180: \hat{\vec{n}} =
181: \frac{\vec{k}_i\times\vec{k}_f}{|\vec{k}_i\times\vec{k}_f|}\,,~~~~
182: \hat{\vec{\ell}} =
183: \frac{\vec{k}_f + \vec{k}_i}{|\vec{k}_f+\vec{k}_i|}\,,~~~~
184: \hat{\vec{m}} =
185: \frac{\vec{k}_f - \vec{k}_i}{|\vec{k}_f-\vec{k}_i|}\,,
186: \end{equation}
187: which satisfy $\hat{\vec{m}}=\hat{\vec{n}}\times\hat{\vec{\ell}}$.
188:
189: For later purposes, it is convenient to choose the invariant
190: normalisation, where the unpolarised cross section is given by%
191: \begin{equation}
192: \left(\frac{\dd\sigma}{\dd t}\right)_{\!np\to np} =
193: \half\left(|a|^2+|b|^2+|c|^2+|d|^2+|e|^2\right),
194: \end{equation}
195: where $t$ is the four-momentum transfer between the initial and
196: final neutrons.
197:
198: In the forward direction $e=0$ and, since one can then not
199: distinguish between the two perpendicular axes,
200: $a(0)-b(0)=c(0)+d(0)$. The scattering matrix then reduces to %
201: \begin{eqnarray}%
202: \nonumber M(\vec{k}_i,\vec{k}_i) &=& \half\big[(a(0) + b(0)) +
203: (c(0)-d(0))(\vec{\sigma}_n\cdot\hat{\vec{\ell}})(\vec{\sigma}_p\cdot
204: \hat{\vec{\ell}})\\ \nonumber &&\hspace{-1.5cm} +(a(0) -b(0))
205: \left\{(\vec{\sigma}_n\cdot\hat{\vec{n}})(\vec{\sigma}_p\cdot\hat{\vec{n}})
206: +(\vec{\sigma}_n\cdot\hat{\vec{m}})(\vec{\sigma}_p\cdot\hat{\vec{m}})\right\}
207: \big]\\ %
208: \end{eqnarray}%
209:
210: There are three spin-correlated total cross sections defined by%
211: \begin{equation}
212: \sigma_{\rm tot}=\sigma_0 -\half\Delta\sigma_{L}\, \mathcal{P}_n^L
213: \mathcal{P}_p^L - \half\Delta\sigma_T\, \vec{\mathcal{P}}_n^T\cdot
214: \vec{\mathcal{P}}_p^T\,,
215: \end{equation}
216: where $\mathcal{P}^L$ and $\vec{\mathcal{P}}^T$ are the
217: longitudinal and transverse components of the polarisation of
218: either the initial neutron or proton.
219:
220: The imaginary parts of the three independent forward amplitudes
221: can be determined through measurements of these total cross
222: sections using the relations:
223: \begin{eqnarray}
224: \nonumber\sigma_0 &=& 2\sqrt{\pi}\,\textit{Im}[a(0)+b(0)]\,,\\
225: \nonumber -\Delta\sigma_{T} &=& 4\sqrt{\pi}\,\textit{Im}[a(0)-b(0)]\,,\\
226: -\Delta\sigma_L &=& 4\sqrt{\pi}\,\textit{Im}[c(0)-d(0)]\,.
227: \end{eqnarray}
228:
229: We are interested in backward rather than forward neutron-proton
230: scattering, in a region that is often called neutron-proton charge
231: exchange. The interchange of the momenta of the final neutron and
232: proton is achieved by letting $\vec{k}_f \to -\vec{k}_f$ and, as
233: is seen from Eq.(\ref{basis1}), this introduces a new set of basis
234: vectors, which are related to the original ones through
235: \begin{equation}\label{basis2}
236: \hat{\vec{n}}_{ce} = - \hat{\vec{n}}\,,~~~\hat{\vec{\ell}}_{ce} =
237: - \hat{\vec{m}}\,,~~~\hat{\vec{m}}_{ce} = - \hat{\vec{\ell}}\,.
238: \end{equation}
239:
240: The scattering matrix in this representation becomes
241: \begin{eqnarray}%
242: \nonumber M(-\vec{k}_f,\vec{k}_i) &=& \half\big[(a + b) + (a -
243: b)(\vec{\sigma}_n\cdot\hat{\vec{n}}_{ce})(\vec{\sigma}_p\cdot\hat{\vec{n}}_{ce})\\
244: \nonumber
245: &&\hspace{-2cm}+(c+d)(\vec{\sigma}_n\cdot\hat{\vec{\ell}}_{ce})
246: (\vec{\sigma}_p\cdot\hat{\vec{\ell}}_{ce})
247: +(c-d)(\vec{\sigma}_n\cdot\hat{\vec{m}}_{ce})(\vec{\sigma}_p\cdot
248: \hat{\vec{m}}_{ce})\\
249: &&-e(\vec{\sigma}_n + \vec{\sigma}_p)\cdot\hat{\vec{n}}_{ce}\big]\,,%
250: \end{eqnarray}%
251: which, in the strictly backward direction of $\theta=\pi$, reduces
252: to
253: \begin{eqnarray}%
254: \nonumber M(-\vec{k}_i,\vec{k}_i) &=& \half\big[(a(\pi) +
255: b(\pi))+\\
256: \nonumber &&\hspace{-2.3cm}
257: (c(\pi)+d(\pi))(\vec{\sigma}_n\cdot\hat{\vec{\ell}}_{ce})(\vec{\sigma}_p\cdot
258: \hat{\vec{\ell}}_{ce})+\\ &&\hspace{-2.3cm}(a(\pi) -b(\pi))
259: \left\{(\vec{\sigma}_n\cdot\hat{\vec{n}}_{ce})(\vec{\sigma}_p\cdot\hat{\vec{n}}_{ce})\right.\nonumber\\
260: &&\hspace{-0.5cm}\left.+(\vec{\sigma}_n\cdot\hat{\vec{m}}_{ce})(\vec{\sigma}_p\cdot\hat{\vec{m}}_{ce})\right\}
261: \big]\,, %
262: \end{eqnarray}%
263: where $e(\pi)=0$ and $a(\pi)-b(\pi)=c(\pi)-d(\pi)$.
264:
265: If one invokes the symmetry properties of the amplitudes that
266: follow from isospin invariance~\cite{BYS78}, the backward values
267: of the imaginary parts of the three independent amplitudes are, in
268: principle, determined by the values of the spin-dependent total
269: cross sections for neutron-proton and proton-proton scattering:
270: \begin{eqnarray}
271: \nonumber
272: \textit{Im}[a(\pi)]&=&\phantom{-}\frac{1}{8\sqrt{\pi}}\left(2\sigma_0^{(-)}
273: -\Delta\sigma_T^{(-)}\right),\\
274: \nonumber
275: \textit{Im}[b(\pi)]&=&\phantom{-}\frac{1}{8\sqrt{\pi}}\left(\Delta\sigma_T^{(-)}
276: +\Delta\sigma_L^{(-)}\right),\\
277: \nonumber
278: \textit{Im}[c(\pi)]&=&\phantom{-}\frac{1}{8\sqrt{\pi}}\left(2\sigma_0^{(-)}
279: +\Delta\sigma_T^{(-)}\right),\\
280: \textit{Im}[d(\pi)]&=&-\frac{1}{8\sqrt{\pi}}\left(\Delta\sigma_T^{(-)}
281: -\Delta\sigma_L^{(-)}\right),
282: \end{eqnarray}
283: where we use the notation
284: \begin{equation}
285: \sigma^{(-)} \equiv \sigma(np)-\sigma(pp)
286: \end{equation}
287: for all three total cross sections.
288:
289: However, in the theoretical treatment of the charge exchange on
290: the deuteron, \textit{i.e.}\ $nd\to p\{nn\}$ at small angles
291: between the incident neutron and final proton, it is convenient to
292: work with an alternative amplitude decomposition~\cite{BW}:
293: \begin{eqnarray}
294: \nonumber
295: M^{ce}(\vec{k}_f,\vec{k}_i)&=&\mathcal{P}_{p\leftrightarrow
296: n}\big[\alpha^{ce} +i\gamma^{ce}
297: (\vec{\sigma}_{n}+\vec{\sigma}_{p})\cdot\hat{\vec{n}}_{ce}\\
298: \nonumber &&\hspace{-2.5cm}+\beta^{ce} (\vec{\sigma}_{n} \cdot
299: \hat{\vec{n}}_{ce})(\vec{\sigma}_{p} \cdot\hat{\vec{n}}_{ce})+
300: \delta^{ce} (\vec{\sigma}_{n} \cdot
301: \hat{\vec{m}}_{ce})(\vec{\sigma}_{p} \cdot \hat{\vec{m}}_{ce})\\
302: \label{Mce} &&\hspace{-2.5cm}+\epsilon (\vec{\sigma}_{n} \cdot
303: \hat{\vec{\ell}}_{ce})(\vec{\sigma}_{p} \cdot
304: \hat{\vec{\ell}}_{ce})\big],
305: \end{eqnarray}
306: where the operator $\mathcal{P}_{p\leftrightarrow n}$ interchanges
307: the charge labels on the final proton and neutron. The presence of
308: this operator means that $\alpha^{ce}$ represents the
309: spin-independent amplitude between the initial neutron and final
310: proton whereas the $(a+b)$ of Eq.~(\ref{BYS}) corresponds to the
311: spin-independent amplitude between the initial and final neutrons.
312:
313: It is straightforward to find the relationship between these two
314: representations and, for the collinear situation that is of
315: interest to us here, we have
316: \begin{eqnarray}
317: \nonumber \alpha^{ce}(0)&=&\fmn{1}{2}\left(a(\pi)+c(\pi)\right)\,,\\
318: \nonumber \beta^{ce}(0)=\delta^{ce}(0)&=&\fmn{1}{2}\left(a(\pi)-c(\pi)\right)\,,\\
319: \epsilon^{ce}(0)&=&\fmn{1}{2}\left(b(\pi)+d(\pi)\right)\,.
320: \label{identification2}
321: \end{eqnarray}
322:
323: The relation between the imaginary parts of these forward
324: amplitudes and the total cross sections is more intuitively
325: obvious:
326: \begin{eqnarray}
327: \nonumber
328: \textit{Im}[\alpha^{ce}(0)]&=&\phantom{-}\frac{1}{4\sqrt{\pi}}\,\sigma_0^{(-)},\\
329: \nonumber
330: \textit{Im}[\beta^{ce}(0))]&=&-\frac{1}{8\sqrt{\pi}}\,\Delta\sigma_T^{(-)},\\
331: \label{sigtot2}
332: \textit{Im}[\epsilon^{ce}(0)]&=&\phantom{-}\frac{1}{8\sqrt{\pi}}\,
333: \Delta\sigma_L^{(-)}.
334: \end{eqnarray}
335:
336: Extra information on the real parts of these amplitudes might be
337: obtained through the use of forward dispersion relations. This is
338: likely to be of most value for the spin-independent term, which
339: only involves unpolarised total cross section input. However, this
340: approach will not be pursued here.
341:
342: From the relations given in Ref.~\cite{BYS78}, the magnitudes of
343: these charge-exchange amplitudes in the forward direction are
344: given in terms of the backward elastic $np$ differential cross
345: section and spin-transfer parameters $K_{0nn0}$ and $K_{0ll0}$
346: through
347: \begin{eqnarray}
348: \nonumber
349: |\alpha^{ce}(0)|^2&=&\fmn{1}{4}\left[1+2K_{0nn0}(\pi)+K_{0ll0}(\pi)\right]
350: \left(\frac{\dd\sigma}{\dd t}\right)_{\!np\to np},\\
351: \nonumber
352: |\beta^{ce}(0)|^2&=&\fmn{1}{4}\left[1-K_{0ll0}(\pi)\right]
353: \left(\frac{\dd\sigma}{\dd t}\right)_{\!np\to np},\\
354: |\epsilon^{ce}(0)|^2&=&\fmn{1}{4}\left[1-2K_{0nn0}(\pi)+K_{0ll0}(\pi)\right]
355: \left(\frac{\dd\sigma}{\dd t}\right)_{\!np\to np}\!\!.
356: \label{observables1}
357: \end{eqnarray}
358: It should be noted that the corresponding results for the
359: non-charge-exchange amplitudes, \emph{i.e.}\ without the
360: interchange operator $\mathcal{P}_{p\leftrightarrow n}$ in
361: Eq.~(\ref{Mce}), have a similar structure but with the $K$ being
362: replaced by the depolarisation parameters $D$. This arises because
363: of the different assignment of the labels \emph{scattered} and
364: \emph{recoil} to the final particles in the two decompositions. Of
365: course, the \emph{elastic} and \emph{charge-exchange}
366: representations must lead to the same physics for elastic
367: neutron-proton scattering, with merely interpretational
368: differences.
369:
370: Using the above relations in association with the phase shift
371: predictions from the SAID database~\cite{SAID}, one sees that the
372: spin-independent term $|\alpha^{ce}(0)|^2$ should contribute less
373: than 10\% to the forward charge-exchange cross section between say
374: 200\,MeV up to the limit of the SAID analysis at 1.3\,GeV. In
375: contrast, there is relatively little spin flip between the initial
376: and final neutrons, \textit{i.e.}\ the $|a(\pi)+b(\pi)|^2$ term
377: dominates. One has therefore to be very careful to specify clearly
378: the meaning of any statement comparing the magnitudes of the
379: spin-flip and spin-independent contributions in backward
380: neutron-proton elastic scattering.
381: %
382: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
383: %
384: \section{Charge exchange on the deuteron}
385: \label{deuteron} \setcounter{equation}{0}
386:
387: In single-scattering (impulse) approximation, the charge exchange
388: $nd\to p\,\{nn\}$ reaction on the deuteron is thought of as a
389: $np\to pn$ reaction with a spectator neutron\footnote{Since the
390: deuteron has isospin-zero, the description of the $pd\to
391: n\,\{pp\}$ reaction is formally identical.}. Initially the
392: neutron-proton pair is bound in the deuteron and the two emerging
393: neutrons are subject to a final state interaction, as illustrated
394: diagramatically in Fig.~\ref{diagram}. If $\vec{k}$, the relative
395: momentum in the $nn$ system, and hence the excitation energy
396: $E_{nn}=k^2/m$, are small, the final neutrons are in the $^{1\!}S_0$
397: state. The reaction therefore acts as a spin-isospin filter going
398: from the $(^{3\!}S_1,^{3\!\!}D_1)$ of the deuteron to the final
399: $^{1\!}S_0$ of the dineutron. Furthermore, at low momentum
400: transfers $\vec{q}=\vec{k}_f-\vec{k}_i$ between the initial
401: neutron and final proton other final states are only weakly
402: excited. Under such conditions the $nd\to p\,\{nn\}$ differential
403: cross section depends but weakly upon the spin-independent
404: amplitude $\alpha^{ce}$ of Eq.~(\ref{Mce}).
405:
406: \begin{figure}[htb]
407: \begin{center}
408: \centerline{\epsfxsize=6cm{\epsfbox{fig1.eps}}}
409: \end{center}
410: \vspace{-0.2cm} \caption{\label{diagram} Impulse approximation
411: diagram for nucleon charge exchange on the deuteron.}
412: \end{figure}
413:
414: The above features were put on a simple quantitative basis through
415: the use of a sum rule by Dean~\cite{Dean} and this was extended to
416: polarisation observables by Bugg and Wilkin~\cite{BW}. The matrix
417: element of the transition is of the form~\cite{Carbonell}
418: \begin{eqnarray}
419: \label{impulse}%
420: \lefteqn{\mathcal{F}(\vec{k}_f,\vec{k}_i;S,\nu_f,M,m_p,m_n) =}\\
421: \nonumber
422: &&\hspace{-4mm}\langle\Psi_{nn,\vec{k}}^{(-)};S,\nu_f,m_p|M^{ce}
423: (\vec{k}_f,\vec{k}_i)
424: \exp(\half i\vec{q}\cdot\vec{r})|\Phi_d^M,m_n\rangle\,,
425: \end{eqnarray}
426: where $\Phi_d(\vec{r})$ is the deuteron wave function in
427: configuration space and $\Psi_{nn,\vec{k}}^{(-)}(\vec{r})$ the
428: corresponding for the low energy $nn$ final state of spin $S$ and
429: projection $\nu_f$. The magnetic quantum numbers of the initial
430: neutron and deuteron and the final proton are denoted by $m_n$,
431: $M$, and $m_p$, respectively.
432:
433: The unpolarised cross section is normalised such that
434: \begin{equation}
435: \label{normd} \frac{\dd4\sigma}{\dd{t}\,\dd^3k}=\fmn{1}{6}\,
436: \textrm{Tr}\left\{\mathcal{F}^{\dagger}\mathcal{F}\right\},
437: \end{equation}
438: where the trace is over all the spin projections.
439:
440: Dean noted that, if one summed Eq.~(\ref{normd}) over all
441: excitation energies of the dineutron, one could obtain a sum rule
442: that did not depend upon the details of the low energy $nn$
443: interaction. This is a high energy approximation and, for it to be
444: valid, the available phase space must be so large as not to
445: disturb the convergence of the sum rule, as discussed below.
446:
447: Specialising to the case of interest here, \emph{viz.},
448: $\theta_{np}=0^{\circ}$, $q\approx 0$, the sum rule for the
449: differential cross section reduces to
450: \begin{eqnarray}
451: \nonumber \lefteqn{\frac{\dd\sigma}{\dd{t}}(nd\to p\{nn\})=\int
452: \frac{\dd^4\sigma}{\dd{t}}\,\dd^3k}\\ &&=
453: \fmn{2}{3}\left(2|\beta^{ce}(0)|^2+|\epsilon^{ce}(0)|^2\right).
454: \end{eqnarray}
455:
456: For completeness, we give also the corresponding sum rule for the
457: deuteron tensor analysing power $t_{20}$;
458: \begin{eqnarray}
459: \nonumber \lefteqn{t_{20}\,\frac{\dd\sigma}{\dd{t}}(nd\to
460: p\{nn\})}\\&&= \frac{2\sqrt{2}}{3}\,\left(1-\fmn{9}{10}P_D\right)
461: \left(|\beta^{ce}(0)|^2-|\epsilon^{ce}(0)|^2\right),
462: \label{tensorrule}
463: \end{eqnarray}
464: where $P_D$ is the deuteron $D$-state probability.
465:
466: The Dean sum rule may then be written in terms of $np\to np$
467: elastic scattering observables as
468: \begin{eqnarray}
469: \nonumber R_{np}(0)&=&\left.\frac{\dd\sigma(nd\to
470: p\{nn\})/\dd{t}}{\dd\sigma(np\to pn)/\dd{t}}\right|_{q=0}\\
471: \nonumber&&\hspace{-1cm}= \frac{2}{3}
472: \frac{2|\beta^{ce}(0)|^2+|\epsilon^{ce}(0)|^2}
473: {|\alpha^{ce}(0)|^2+2|\beta^{ce}(0)|^2+|\epsilon^{ce}(0)|^2}\\
474: &&\hspace{-1cm}=
475: \fmn{1}{6}\left[3-K_{0ll0}(\pi)-2K_{0nn0}(\pi)\right].
476: \label{Rnptheory}
477: \end{eqnarray}
478:
479: The sum rule of Eq.~(\ref{Rnptheory}) is very effective at medium
480: and high energies because it converges so quickly. This is
481: illustrated in Fig.~\ref{sumrulebye}, where the impulse
482: approximation cross section of Eq.~(\ref{normd}) has been
483: evaluated for the $pd\to n\{pp\}$ case~\cite{Carbonell} and
484: integrated numerically over the $pp$ excitation energy. Although a
485: specific nucleon-nucleon potential was used~\cite{Sprung}, the
486: rate of convergence depends little upon this choice and less than
487: 1\% of the sum rule remains beyond $E_{pp}\approx 15$\,MeV. The
488: rate of convergence under the specific conditions of the Dubna
489: $d(n,p)nn$ experiment~\cite{Sharov08} has also been
490: discussed by Ladygina~\cite{Ladygina}.
491:
492: \begin{figure}[htb]
493: \begin{center}
494: \centerline{\epsfxsize=8cm{\epsfbox{fig2.eps}}}
495: \end{center}
496: \vspace{-0.2cm} \caption{Convergence of the sum rule for the
497: $pd\to n\{pp\}$ reaction at $q=0$ as a function of the excitation
498: energy in the final $pp$ system. It has been assumed that the
499: spin-non-flip amplitude $\alpha^{ce}=0$ to make the limiting value
500: $\frac{2}{3}$, though the rate of convergence is similar in all
501: cases. The evaluation has been carried out using the de Toureill
502: and Sprung potential~\cite{Sprung} to describe the low energy
503: nucleon-nucleon systems, as explained in Ref.~\cite{Carbonell}.
504: \label{sumrulebye}}
505: \end{figure}
506:
507: Deviations might be expected from the sum rule at lower energies
508: where the phase space available does not allow an unimpeded
509: integration over $E_{nn}$. As discussed in section~\ref{data},
510: this has only a marginal effect on the saturation of the sum rule
511: for incident energies above 50\,MeV. More critical, though, is the
512: fact that at low energies experimentalists generally put a more
513: severe cut on the momentum of the final proton in the $nd\to
514: p\{nn\}$ reaction to minimise the contributions from diagrams
515: other than those of the impulse approximation and this reduces the
516: value obtained for $R_{np}(0)$. Although we later report results
517: at low energies, there is no reason to believe that the impulse
518: approximation of Fig.~\ref{diagram} should dominate there and a
519: full three-body calculation has to be undertaken to interpret
520: these results.
521:
522: The most significant correction to the sum rule at high energies
523: comes from the shadow effect~\cite{Glauber}, which will typically
524: reduce the value of $R_{np}(0)$ by about 5\%.
525: %
526: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
527: %
528: \section{Experimental data and theoretical comparison}
529: \label{data} \setcounter{equation}{0}
530:
531: \begin{table}[hbt]
532: \begin{center}
533: \caption{Summary of the available experimental data on the
534: $R_{np}(0)$ ratio measured using the $nd\to p\,\{nn\}$ reaction.
535: The error bars reflect both the statistical and systematic
536: uncertainties.\label{table1}}
537: \begin{tabular}{|r|c|c|r|}
538: \hline
539: &&&\\
540: $T_{\rm kin}$~~ & $R_{np}(0)$ & Year & Ref.\\
541: (MeV) & & & \\
542: \hline
543: 13.9 & $0.185~~~~~~~~~~~$ & 1965 &\cite{VOI65} \\
544: %
545: 90.0 & $0.397 \pm 0.044$ & 1951 &\cite{POW51} \\
546: %
547: 152.0 & $0.650 \pm 0.100$ & 1966 &\cite{MEA66} \\
548: %
549: 200.0 & $0.553 \pm 0.030$ & 1962 &\cite{DZH62} \\
550: %
551: 270.0 & $0.710 \pm 0.021$ & 1952 &\cite{CLA52} \\
552: %
553: 299.7 & $0.652 \pm 0.033$ & 1988 &\cite{PAG88} \\
554: %
555: 319.8 & $0.643 \pm 0.032$ & 1988 &\cite{PAG88} \\
556: %
557: 339.7 & $0.637 \pm 0.032$ & 1988 &\cite{PAG88} \\
558: %
559: 359.6 & $0.626 \pm 0.031$ & 1988 &\cite{PAG88} \\
560: %
561: 379.6 & $0.641 \pm 0.032$ & 1988 &\cite{PAG88} \\
562: %
563: 380.0 & $0.200 \pm 0.035$ & 1955 &\cite{DZH55} \\
564: %
565: 399.7 & $0.610 \pm 0.031$ & 1988 &\cite{PAG88} \\
566: %
567: 419.8 & $0.623 \pm 0.031$ & 1988 &\cite{PAG88} \\
568: %
569: 440.0 & $0.630 \pm 0.032$ & 1988 &\cite{PAG88} \\
570: %
571: 460.1 & $0.611 \pm 0.031$ & 1988 &\cite{PAG88} \\
572: %
573: 480.4 & $0.608 \pm 0.030$ & 1988 &\cite{PAG88} \\
574: %
575: 500.9 & $0.592 \pm 0.030$ & 1988 &\cite{PAG88} \\
576: %
577: 521.1 & $0.604 \pm 0.030$ & 1988 &\cite{PAG88} \\
578: %
579: 539.4 & $0.617 \pm 0.031$ & 1988 &\cite{PAG88} \\
580: %
581: 550.0 & $0.589 \pm 0.046$ & 2007 &\cite{Sharov08} \\
582: %
583: 557.4 & $0.632 \pm 0.032$ & 1988 &\cite{PAG88} \\
584: %
585: 710.0 & $0.483 \pm 0.080$ & 1960 &\cite{LAR60} \\
586: %
587: 794.0 & $0.560 \pm 0.040$ & 1978 & \cite{BON78} \\
588: %
589: 800.0 & $0.554 \pm 0.023$ & 2007 & \cite{Sharov08} \\
590: %
591: 1000\phantom{.0} & $0.553 \pm 0.026$ & 2007 & \cite{Sharov08} \\
592: %
593: 1200\phantom{.0} & $0.551 \pm 0.022$ & 2007 & \cite{Sharov08} \\
594: %
595: 1400\phantom{.0} & $0.576 \pm 0.038$ & 2007 & \cite{Sharov08} \\
596: %
597: 1800\phantom{.0} & $0.568 \pm 0.033$ & 2007 & \cite{Sharov08} \\
598: %
599: 2000\phantom{.0} & $0.564 \pm 0.045$ & 2007 & \cite{Sharov08} \\
600: \hline
601:
602: \end{tabular}
603: \end{center}
604: \end{table}
605:
606:
607:
608: %
609: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
610: %
611: \begin{table}[hbt]
612: \begin{center}
613: \caption{Summary of the available experimental data on the
614: $R_{np}(0)$ ratio measured using the $pd\to n\,\{pp\}$ reaction.
615: The error bars reflect both the statistical and systematic
616: uncertainties. \label{table2}}
617: \end{center}
618: \begin{center}
619: \begin{tabular}{|r|c|c|c|}
620: \hline
621: &&&\\
622: $T_{\rm kin}$~~ & $R_{np}(0)$ & Year & Ref.\\
623: (MeV) & & & \\
624: \hline
625: ~13.5 & $0.180~~~~~~~~~~~$ & 1959 &\cite{WON59} \\
626: %\hline
627: ~30.1 & $0.141 \pm 0.035$ & 1967 &\cite{BAT66} \\
628: %\hline
629: ~50.0 & $0.240 \pm 0.060$ & 1967 &\cite{BAT66} \\
630: %\hline
631: ~95.0 & $0.480 \pm 0.030$ & 1953 &\cite{HOF53} \\
632: %\hline
633: ~95.7 & $0.587 \pm 0.029$ & 1967 &\cite{LAN67} \\
634: %\hline
635: 135.0 & $0.652 \pm 0.154$ & 1965 &\cite{EST65} \\
636: %\hline
637: 143.9 & $0.601 \pm 0.057$ & 1967 &\cite{LAN67} \\
638: %\hline
639: 647.0 & $0.600 \pm 0.080$ & 1976 &\cite{BJO76} \\
640: %\hline
641: 800.0 & $0.660 \pm 0.080$ & 1976 & \cite{BJO76} \\
642: \hline
643: \end{tabular}
644: \end{center}
645: \end{table}
646:
647: \begin{table}[hbt]
648: \begin{center}
649: \caption{Summary of the available experimental data on the
650: $R_{np}(0)$ ratio measured using the $dp\to \{pp\}\,n$ reaction.
651: The kinetic energy quoted here is the energy per nucleon. The
652: error bars reflect both the statistical and systematic
653: uncertainties. \label{table3}}
654: \begin{tabular}{|c|c|c|c|}
655: \hline
656: &&& \\
657: $T_{\rm kin}$ & $R_{np}(0)$ & Year & Ref.\\
658: (MeV) & & & \\
659: \hline
660: 977.0 & $0.430 \pm 0.220$ & 1975 & \cite{ALA75} \\
661: 977.0 & $0.650 \pm 0.120$ & 2002 & \cite{GLA02} \\
662: \hline
663: \end{tabular}
664: \end{center}
665: \end{table}
666: %
667: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
668: %
669:
670: The cross section ratio $R_{np}(0)$ can be investigated using
671: either the $nd \to p\{nn\}$, $pd\to n\{pp\}$, or the $dp\to \{pp\}
672: n$ reaction and the experimental results in the three cases are
673: summarised, respectively, in Tables~\ref{table1}, \ref{table2},
674: and \ref{table3}. In many cases the results had to be read from
675: graphs and this is especially true of some of the very old data.
676: However, because of their age, such data tend to be at the lower
677: energies, which are of less interest for the amplitude
678: reconstruction.
679:
680: There are more measurements of the $nd \to p\{nn\}$ reaction than
681: of $pd\to n\{pp\}$ and the error bars are also generally smaller.
682: This reflects the relative difficulty in the two types of
683: experiment. If a good quality neutron beam is available, the
684: measurement of the relative strengths of the proton spectra from
685: deuterium and hydrogen targets is comparatively
686: \textit{straightforward}. The alternative of measuring the neutron
687: in the final state presents far more difficulties. The third
688: technique of using a deuteron beam has only been attempted once
689: because in this case one has to measure both final fast protons
690: over a large range of phase space. This can be done using a bubble
691: chamber and the value~\cite{GLA02} quoted in Table~\ref{table3}
692: represents the result of increased statistics and a different
693: analysis compared to that of the previous
694: publication~\cite{ALA75}. A particular criticism here is that the
695: free neutron-proton data have to be taken from an entirely
696: different source so that there can be no cancellations between any
697: systematic errors.
698:
699: Although, as shown by Fig.~\ref{sumrulebye}, the sum rule
700: converges quite fast, there is nevertheless a problem at low
701: energies as to where to cut the tail in the proton spectrum and, in
702: certain cases, it is likely that the choice has not allowed for a
703: full saturation of the sum rule.
704:
705: \begin{figure}[hbt]
706: \centering
707: \includegraphics[width=0.9\columnwidth,clip]{fig3.eps}
708: \caption{Experimental data on the $R_{np}(0)$ ratio at zero
709: momentum transfer. The closed circles are from the $(n,p)$ data of
710: Table~\ref{table1}, the open circles from the $(p,n)$ data of
711: Table~\ref{table2}, and the cross from the $(d,2p)$ datum of
712: Table~\ref{table3}. These results are compared to the Dean
713: sum-rule predictions of Eq.~(\ref{Rnptheory}) using the current
714: SAID solution, which is available up to a laboratory kinetic
715: energy of 1.3\,GeV. The dashed curve takes into account the
716: limited phase space available at the lower
717: energies.\label{sumrule}}
718: \end{figure}
719:
720: The different measurements of $R_{np}(0)$ are reported in
721: Fig.~\ref{sumrule} with one controversial point~\cite {DZH55}
722: being omitted. Also shown are the estimates of
723: Eq.~(\ref{Rnptheory}), evaluated using the current SAID
724: solution~\cite{SAID}. In these predictions, no account has been
725: taken of the shadow effect~\cite{Glauber}, which would reduce the
726: limiting value to typically 0.63/0.64. The agreement with the
727: phase shift predictions is generally satisfactory, though there
728: are suggestions from the new data~\cite{Sharov08} that there might
729: might be some underestimate of the relative strength of the
730: $|\alpha^{ce}(0)|^2$ contribution at the higher energies.
731:
732: At low energies the predictions fall steeply. In part this is
733: related to the behaviour of the $np$ amplitudes but one has also
734: to take account of the fact that the smaller available phase space
735: does not allow the sum rule to be fully saturated. However, even
736: when this effect is included, by integrating numerically the
737: impulse approximation, the effect is comparatively small and does
738: not account for the much steeper drop in the experimental data. Of
739: much more importance there is the experimental cut in the recoil
740: proton momentum and, of course, the deviations from the naive
741: impulse approximation at low energies.
742:
743: %
744: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
745: %
746: \section{Summary and conclusions}
747: \label{summary} \setcounter{equation}{0}
748:
749: The ratio $R_{np}(0)$ of the forward differential cross section
750: for charge exchange on the deuteron to that on the nucleon is a
751: very robust observable since many uncertainties drop out between
752: the two measurements. It is therefore not surprising that one gets
753: a consistent picture of the energy dependence of this quantity
754: from measurements of the $nd \to p\{nn\}$, $pd\to n\{pp\}$, and
755: $dp\to \{pp\} n$ reactions. Furthermore, we have shown that, from
756: about 300\,MeV up to 1.3\,GeV, where the $np$ phase shift analyses
757: terminate, the predictions for $R_{np}(0)$ largely agree with the
758: available experimental data. Since measurements of $R_{np}(0)$
759: have now been carried out up to 2\,GeV~\cite{Sharov08}, it is
760: appropriate to consider whether a neutron-proton elastic amplitude
761: reconstruction could be performed using these results.
762:
763: In the backward direction, we see from Eq.~(\ref{identification2})
764: that there are only three (complex) amplitudes, $\alpha^{ce}(0)$,
765: $\beta^{ce}(0)$, and $\epsilon^{ce}(0)$. In principle, as shown by
766: Eq.~(\ref{sigtot2}), the imaginary parts of these amplitudes can
767: be fixed by measurements of the spin dependence of the $pp$ and
768: $np$ total cross sections. According to the available phase shift
769: analyses, these quantities are much smaller than the real parts
770: and so the error bars will be relatively large, even if all the
771: measurements were available. The values of $|\alpha^{ce}(0)|^2$
772: and $2|\beta^{ce}(0)|^2+|\epsilon^{ce}(0)|^2$ would be fixed by
773: the combined measurement of $R_{np}(0)$ and the free $np\to np$
774: differential cross section.
775:
776: Measurements have been carried out on the $t_{20}$ tensor
777: analysing power of the
778: $d\hspace{-3.2mm}\stackrel{\to}{\phantom{p}}\hspace{-1.4mm}p\to
779: \{pp\}n$ reaction~\cite{ELL87,KOX93,CHI06}. Rather than using the
780: sum rule of Eq.~(\ref{tensorrule}), these data were taken with an
781: excitation energy in the $pp$ system so low that the final
782: $^{1\!}S_0$ system dominates or where one could correct from
783: contamination for the $pp$ $P$-waves. Under such conditions,
784: \begin{equation}
785: t_{20}=\sqrt{2}\left(\frac{|\beta^{ce}(0)|^2-|\epsilon^{ce}(0)|^2}
786: {2|\beta^{ce}(0)|^2+|\epsilon^{ce}(0)|^2}\right)\cdot
787: \end{equation}
788: Furthermore, a measurement of the transverse spin correlation with
789: an incident vector polarised deuteron on a polarised hydrogen
790: target would give~\cite{Barbaro}
791: \begin{equation}
792: C_{n,n}(0)=\frac{-2\textit{Re}(\beta^{ce}(0)^*\epsilon^{ce}(0))}
793: {2|\beta^{ce}(0)|^2+|\epsilon^{ce}(0)|^2}\cdot
794: \end{equation}
795:
796: It is therefore clear that, even with very good data on these
797: seven parameters, there would still be two discrete ambiguities,
798: \textit{viz.}\ the signs of the $\textit{Re}(\alpha^{ce}(0)$ and
799: $\textit{Re}(\beta^{ce}(0)$. However, since there are extensive
800: data on the unpolarised cross section difference of
801: Eq.~(\ref{sigtot2}), the application of forward dispersion
802: relations will certainly be sufficient to determine at least the
803: sign of $\textit{Re}(\alpha^{ce}(0)$. Regarding the other
804: ambiguity, if $\textit{Re}(\beta^{ce}(0)$ changes sign then this
805: would be reflected in the value of $t_{20}(0)$, which has been
806: found to have the same sign throughout the measured
807: range~\cite{ELL87,KOX93,CHI06}. In principle, therefore, it seems
808: that a direct amplitude construction of $np\to np$ in the backward
809: direction is feasible without having to measure differential cross
810: sections with polarised protons and neutrons, although these would
811: clearly enhance the precision of any analysis.
812:
813: For obvious experimental reasons, the measurements of deuteron
814: charge exchange leading to the low excitation energy $pp$ system
815: have so far been carried out with a polarised deuteron
816: beam~\cite{ELL87,KOX93,CHI06}. This lowers the limit on the energy
817: per nucleon. However, a polarised hydrogen/deuterium gas target is
818: now available at the ANKE facility of the COSY-J\"ulich storage
819: ring~\cite{SPIN}. Using solid state telescopes placed within the
820: target chamber, the low energy protons from the
821: % Would really like $\vec{p}\vec{d}\to n\{pp\}$ but not allowed by
822: % the EPJA style file
823: $p\hspace{-3mm}\stackrel{\to}{\phantom{p}}\hspace{-1.2mm}d\hspace{-2.7mm}\stackrel{\to}{\phantom{p}}\to n\{pp\}$
824: reaction can be measured with high precision. This will
825: allow the values of $t_{20}(0)$ and $C_{n,n}(0)$ to be
826: investigated up to the maximum proton beam energy of 2.9\,GeV.
827:
828: %
829: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
830: %
831:
832: \begin{acknowledgement}
833: This work has resulted from an active and productive collaboration
834: with the Dubna DELTA-SIGMA group of Ref.~\cite{Sharov08}, for
835: which we are very appreciative. We are also grateful to J.~Ludwig
836: for furnishing us with a copy of Ref.~\cite{PAG88}. One of the
837: authors (CW) wishes to thank the Czech Technical University, where
838: this work was initiated, for hospitality and support.
839: \end{acknowledgement}
840: %
841: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
842: %
843:
844: \begin{thebibliography}{99}
845: %
846: \bibitem{SAID} R.A.~Arndt, I.I.~Strakovsky, and R.L.~Workman,
847: Phys.\ Rev.\ C \textbf{62} (2000) 034005;
848: \verb=http://gwdac.phys.gwu.edu=.
849: %
850: \bibitem{Ball1} J.~Ball \textit{et al.}, Eur.\ Phys.\ J.\ C
851: \textbf{5} (1998) 57.
852: %
853: \bibitem{Ball2} J.~Ball \textit{et al.}, Eur.\ Phys.\ J.\
854: \textbf{11} (1999) 51.
855: %
856: \bibitem{Lesquen} A.~de~Lesquen \textit{et al.}, Eur.\ Phys.\ J.\
857: \textbf{11} (1999) 69.
858: %
859: \bibitem{Pomeranchuk} I.~Pomeranchuk, Doklady Akad.\ Nauk \textbf{77} (1951)
860: 249.
861: %
862: \bibitem{Dean} N.W.~Dean, Phys.\ Rev.\ D \textbf{5} (1972) 1661;
863: N.W.~Dean, Phys.\ Rev.\ D \textbf{5} (1972) 2832.
864: %
865: \bibitem{Sharov08} V.I.~Sharov \textit{et al.}, Dubna report
866: E1-2008-61 (2008).
867: %
868: \bibitem{BYS78} J.~Bystrick\'y, F.~Lehar, and P.~Winternitz,
869: J.\ Phys.\ (Paris) \textbf{39} (1978) 1.
870: %
871: \bibitem{BW} D.V.~Bugg and C.~Wilkin,
872: Nucl.\ Phys.\ A \textbf{467} (1987) 575.
873: %
874: \bibitem{Carbonell} J~Carbonell, M.B.~Barbaro, and C.~Wilkin,
875: Nucl.\ Phys.\ A \textbf{529} (1991) 653.
876: %
877: \bibitem{Sprung} R.~de Tourreil and D.W.~Sprung, Nucl.\ Phys.\ A
878: \textbf{201} (1973) 193.
879: %
880: \bibitem{Ladygina} N.B.~Ladygina, Phys.\ Atom.\ Nucl.\ \textbf{71}
881: (2007) 58.
882: %
883: \bibitem{Glauber} R.J.~Glauber, in \textit{Lectures in Theoretical
884: Physics}, ed.\ W.E.~Brittin (Interscience, N.Y.\ 1959) vol.~1,
885: p.~315.
886: %
887: \bibitem{VOI65} V.K.~Voitovetskii, I.L.~Korsunskii, and Yu.F.~Pazhin,
888: Nucl.\ Phys.\ \textbf{69} (1965) 531.
889: %
890: \bibitem{POW51} W.M.~Powell, UCRL report 1191 (1951).
891: %
892: \bibitem{MEA66} D.F.~Measday, Phys.\ Lett.\ \textbf{21} (1966) 66.
893: %
894: \bibitem{DZH62} V.P.~Dzhelepov, Proc.\ Int.\ Conf.\ High Energy
895: Phys., Ed.~J.~Prentki, (CERN, Geneva, 1962) p.~19.
896: %
897: \bibitem{CLA52} J.R.~Cladis, J.~Hadley, and W.N.~Hess,
898: Phys.\ Rev.\ \textbf{86} (1952) 110.
899: %
900: \newpage
901: %
902: \bibitem{PAG88} B.~Pagels,
903: Diplomarbeit : \textit{Untersuchung der quasielastischen
904: Ladungsaustauschreaktion $nd \rightarrow pnn$
905: im Neutronenergiebereich von 290 bis 570\,MeV},
906: Universit\"at Freiburg im Breisgau (1988).
907: %
908: \bibitem{DZH55} V.P.~Dzhelepov \textit{et al.},
909: Izvestia Akad.\ Nauk \textbf{19} (1955) 573.
910: %
911: \bibitem{LAR60} R.R.~Larsen, Nuovo Cim.\ \textbf{18} (1960) 1039.
912: %
913: \bibitem{BON78} B.E.~Bonner \textit{et al.}, Phys.\ Rev.\ C
914: \textbf{17} (1978) 664.
915: %
916: \bibitem{WON59} C.~Wong \textit{et al.}, Phys.\ Rev.\
917: \textbf{116} (1959) 104.
918: %
919: \bibitem{BAT66} C.J.~Batty, R.S.~Gilmore and G.H.~Stafford,
920: Phys.\ Lett.\ \textbf{16} (1965) 137;
921: A.S.~Clough \textit{et al.}. Nucl.\ Phys.\ A
922: \textbf{121} (1968) 689.
923: %
924: \bibitem{HOF53} J.A.~Hofmann and K.~Strauch,
925: Phys.\ Rev.\ \textbf{90} (1953) 559.
926: %
927: \bibitem{LAN67} A.~Langsford \textit{et al.},
928: Nucl.\ Phys.\ A \textbf{99} (1967) 246.
929: %
930: \bibitem{EST65} M.J.~Esten, T.C.~Griffith, G.J.~Lush, A.J.~Metheringham,
931: Rev.\ Mod.\ Phys.\ \textbf{37} (1965) 533.
932: %
933: \bibitem{BJO76} C.W.~Bjork \textit{et al.}, Phys.\ Lett.\
934: \textbf{63B} (1976) 31.
935: %
936: \bibitem{ALA75} B.S.~Aladashvili \textit{et al.}, Nucl.\ Phys.\
937: B \textbf{86} (1975) 461.
938: %
939: \bibitem{GLA02} V.V.~Glagolev \textit{et al.}, Eur.\ Phys.\ J.\ A
940: \textbf{15} (2002) 471; \textit{idem}
941: Dubna report P1-2006-112 (2006).
942: %
943: \bibitem{ELL87} C.~Ellegaard \textit{et al.}, Phys.\ Rev.\ Lett.\
944: \textbf{59} (1987) 974; T.~Sams \textit{et al.},
945: Phys.\ Rev.\ C \textbf{51} (1995) 1945.
946: %
947: \bibitem{KOX93} S.~Kox \textit{et al.}, Nucl.\ Phys.\ A
948: \textbf{556} (1993) 621.
949: %
950: \bibitem{CHI06} D.~Chiladze \textit{et al.}, Phys.\ Lett.\ B
951: \textbf{637} (2006) 170.
952: %
953: \bibitem{Barbaro} M.B.~Barbaro and C.~Wilkin, J.\ Phys.\ G \textbf{15} (1989)
954: L69.
955: %
956: \bibitem{SPIN} A.~Kacharava, F.~Rathmann, and C.~Wilkin, COSY Proposal \#152,
957: \textit{Spin Physics from COSY to FAIR},
958: \verb=arXiv:nucl-ex:0511028}=.
959: \end{thebibliography}
960: \end{document}
961: