1:
2: \documentclass{elsart}
3:
4:
5: \usepackage{epsfig}
6:
7:
8: \usepackage{amssymb}
9:
10: \begin{document}
11:
12: \begin{flushright}
13:
14: {\bf DFUB 6/2003}
15:
16: \end{flushright}
17: \begin{frontmatter}
18:
19:
20: \title{Monte Carlo simulation of an experiment
21: looking for radiative solar neutrino decays}
22:
23: \author[bof,tesre]{S. Cecchini}
24: \author[bof]{D. Centomo}
25: \author[bof]{G. Giacomelli}
26: \author[bof,iss]{V. Popa}
27: \author[bof,iss]{C.G. \c{S}erb\u{a}nu\c{t}}
28: \address[bof]{Dipartimento di Fisica dell'Universit\`{a} and INFN Sezione di Bologna,
29: I-40127 Bologna, Italy}
30: \address[tesre]{IASF/CNR, I-40129 Bologna, Italy}
31: \address[iss]{Institute for Space Sciences, R-77125 Bucharest M\u{a}gurele,
32: Romania}
33:
34:
35: \begin{abstract}
36: We analyse the possibility of detecting visible photons from a hypothetical radiative
37: decay of solar neutrinos. Our study is focused on the simulation of such
38: measurements during total solar eclipses and it is based on the BP2000 Standard Solar
39: Model and on the most recent experimental information
40: concerning the neutrino properties. Our calculations yield the
41: probabilities of the decays, the shapes of the visible signals and the spectral
42: distributions of the expected photons, under the assumption that solar neutrino
43: oscillations occur according to the LMA model.
44: \end{abstract}
45:
46: \begin{keyword}
47: Solar neutrinos \sep Decays of heavy neutrinos \sep Neutrino mass and mixing
48: \sep Total solar eclipses \sep Numerical simulations
49:
50: \PACS 96.60.Vg \sep 13.35.Hb \sep 14.60.Pq \sep 95.85.Ry \sep 02.60.Cb
51: \end{keyword}
52: \end{frontmatter}
53:
54:
55: \section{Introduction}
56: \label{intro}
57:
58: In the last few years it has become clear that neutrinos have non-vanishing
59: masses, and that the neutrino flavor eigenstates ($\nu_e$, $\nu_\mu$ and $\nu_\tau$)
60: are superpositions of mass eigenstates ($\nu_1$, $\nu_2$ and $\nu_3$). For a recent
61: review, see \cite{gg}. In this context, neutrinos could
62: undergo radiative decays, e.g. $\nu_2 \rightarrow
63: \nu_1 + \gamma$, as initially suggested in \cite{sciama}. The present
64: status of decaying theory is sumarized in \cite{sciama2}.
65: Such decays request
66: that the involved neutrinos have a non-vanishing electric dipole moment; the
67: very stringent existing experimental limits refer to the flavor neutrino eigenstates
68: and they are not directly applicable to possible dipole moments of mass neutrino
69: eigenstates.
70:
71: In a pioneering experiment performed during the
72: Total Solar Eclipse (TSE) of October 24, 1995
73: a search was made for visible photons emitted through possible radiative decays
74: of solar neutrinos during their flight between the Moon and the Earth \cite{vanucci}.
75: In the analysis
76: of the data, the authors assumed
77: that all neutrinos have masses of the order of
78: few eV, $\Delta m_{12}^2 = m_2^2 - m_1^2
79: \simeq 10^{-5}$ eV$^2$, and an average neutrino energy of
80: 860 keV; furthermore they assumed that all decays would lead to visible photons,
81: which would travel nearly in the same direction as the parent neutrinos, thus
82: leading to a narrow spot of light
83: coming from the direction of
84: the center of the dark disk of the
85: Moon.
86:
87: Subsequently,
88: Fr\`{e}re and Monderen made more accurate calculations on the shape of the expected
89: signal \cite{frere}, considering the Sun as an extended source of electron
90: neutrinos with
91: typical energies of the order of 1 MeV,
92: a Lagrangian formalism for the radiative decay,
93: different neutrino masses (1 eV or 0.5 eV), and mass
94: squared
95: differences $\Delta m^2$ of
96: $10^{-5}$, 0.25 and 1 eV$^2$; they have shown
97: that the expected signal could have an extended angular pattern.
98:
99: Some of the authors of this paper
100: were involved in two experiments along the line of \cite{vanucci}, during
101: the total solar eclipses of August 11, 1999 (in Romania)
102: \cite{n1,n2,n3}, and of June 21, 2001 (in Zambia) \cite{n3}.
103: In 1999 the bad weather conditions did not allow the planned observations, but
104: we could use a videotape filmed by a local television (R\^{a}mnicu V\^{a}lcea).
105: The analysis of
106: the data was performed
107: in the hypothesis of a possible
108: \begin{equation}
109: \label{deca}
110: \nu_2 \rightarrow \nu_1 + \gamma
111: \end{equation}
112: decay, with $m_2 > m_1$.
113: For the analysis of the 1999 data we have chosen
114: the two $\Delta m^2$ values suggested by the
115: MSW SMA (Small Mixing Angle) and LMA (Large Mixing Angle) solutions of the Solar
116: Neutrino Problem (SNP), allowed by the then available experimental data from solar
117: neutrino experiments.
118: We developed a Monte Carlo (MC) simulation of the radiative
119: solar neutrino decay, considering the solar neutrino energy spectrum predicted
120: by the Standard Solar Model (SSM) \cite{bahcall1} and the mass of the $\nu_1$
121: in the range of 1 - 10 eV, as it was expected at that time.
122: Since the angular resolution of the data was not very
123: good,
124: we considered the Sun as a pointlike neutrino surce. The simulation has shown that
125: the expected signal should be a narrow spot of light in the direction of the center
126: of the Sun, and allowed an evaluation of the fraction of decays yielding visible
127: photons as function of the chosen neutrino $\nu_1$ mass $m_1$
128: and $\Delta m^2$ values.
129:
130: The 2001 experiment lead to better quality data, so the real
131: spatial distribution of
132: the solar neutrino yield had to be considered. Furthermore, the recent SNO
133: results \cite{sno1,sno2} favour the LMA solution and
134: could indicate also the presence of a $\nu_\tau$ component in the solar neutrino
135: flux at the Earth level.
136:
137: The WMAP (Wilkinson Microwave Anisotropy Probe)
138: results after the first year of flight
139: \cite{map} limit the sum of the masses of the three
140: neutrino species to 0.23 eV ( 95\%
141: Confidence Level).
142:
143: In this paper we present a more complete simulation.
144: We assume that
145: $m_1 < m_2 < m_3$ where $m_1$, $m_2$, $m_3$ are the masses
146: of the $\nu_1$, $\nu_2$ and
147: $\nu_3$ mass eigenstates, respectively; but we
148: restrict our analysis to a two generation mixing
149: scenario, assuming the present
150: mass differences obtained from solar neutrino experiments,
151: the LMA solution with $\Delta m^2_{12} = 6 \times 10^{-5}$ eV$^2$.
152: Since SNO suggests also
153: the presence of $\nu_3$ in the solar neutrino flux, we considered also the mass
154: difference measured by atmospheric neutrino experiments \cite{macro,miri,sk}:
155: $\Delta m^2_{13} \simeq
156: \Delta m^2_{23} = 2.5 \times 10^{-3}$ eV$^2$. We included in our MC
157: all the details of neutrino production in the Sun, as given by the
158: ``BP2000" SSM \cite{bahcall2}.
159:
160: The aim of our simulation is to give information on the shape of a possible decay
161: signal of solar neutrinos (angular and energy distributions
162: of the emitted photons) and to estimate
163: the probabilities for the decay and the geometrical detection efficiency.
164: This information should allow to obtain limits on
165: the neutrino lifetimes from the experimental observations. As it
166: is shown in the following sections, some of the input parameters of
167: the code refer to the characteristics of a specific experiment. The results
168: presented in this paper are obtained in the conditions of the observations
169: made in 2001 in Zambia \cite{n3} with a digital videocamera.
170:
171: \section{The Monte Carlo simulation}
172: \label{mc}
173:
174: The assumed geometry for the simulation of the radiative decay of solar
175: neutrinos during a TSE is shown in Fig. 1.
176: The notations in the figure will
177: be used in the following subsections.
178:
179:
180: \begin{figure}
181: \vspace{-10mm}
182: \begin{center}
183: \mbox{ \epsfysize=9cm
184: \epsffile{mc.eps}}
185: \caption{A sketch of the geometry of the production
186: of solar neutrinos, their possible radiative decay (in the space between the
187: Moon and the Earth) and the detection of the emitted photon, during a TSE.
188: The $z$ axis is directed from the center of the Sun to the observation point
189: on the Earth (or center of the Earth)}
190: \end{center}
191: \end{figure}
192:
193: \subsection{Neutrino production inside the Sun}
194: \label{soare}
195:
196: In simulating the solar neutrino production, we used the ``BP2000" SSM \cite{bahcall2}
197: in its numerical form directly available from \cite{web}.
198: The first step consists in chosing a specific reaction/decay
199: yielding neutrinos (both
200: from the p-p and the CNO cycles) according to its predicted contribution
201: to the solar neutrino flux at the Earth. The neutrino energy and
202: the radius $R$ of its production point are then randomly generated according
203: to the SSM. In Fig. 2
204: we present the neutrino solar energy spectrum
205: obtained from
206: $3 \times 10^7$ generated neutrinos, and the
207: distribution of the distance $R$ from the center of the Sun of the production points.
208:
209: \begin{figure}
210: %\vspace{-17mm}
211: \begin{center}
212: \mbox{ \epsfysize=7cm
213: \epsffile{fig2.eps}}
214: \caption{Monte Carlo simulations according to the ``BP2000" SSM. (a)The energy
215: spectrum of solar neutrinos; the $^8B$ neutrinos are the few neutrinos with energy
216: between $10^3$ and $10^4$ keV.
217: (b) The distribution of the radial distance of the production points from the center of
218: the Sun.
219: }
220: \end{center}
221: \end{figure}
222:
223:
224: In order to ensure a uniform distribution of the birth points of the neutrinos
225: ($P_1$ in Fig. 1) on each shell $\Delta R$ at radial distance $R$,
226: we determine for each neutrino
227: the angular spherical coordinates
228: generating uniformely the cosinus of the zenithal angle ($\theta_S$ in Fig. 1,
229: assuming the ``z" axis
230: oriented from the centre of the Sun towards the centre of the Earth) and the
231: polar angle ($\phi_S$ in Fig. 1.).
232:
233: \subsection{Propagation of solar neutrinos and of the decay photons}
234: \label{pamant}
235:
236: The solar neutrinos are produced as $\nu_e$'s, thus as a superposition
237: of neutrino mass states. After leaving the Sun, during their flight in the
238: interplanetary space, their weak flavor is irrelevant.
239:
240: In our simulation we
241: consider the $\nu_2$ (or $\nu_3$) component of the neutrino flux, that may decay into
242: the lower mass state $\nu_1$ and a photon. As we are interested in the possibility
243: to observe such photons during a total solar eclipse, we impose that the decay
244: processes take place in the space between the Moon and the Earth,
245: inside the shadow cone of the Moon (otherwise the separation of the decay photons
246: from the solar light backround would be impossible). Furthermore,
247: the decay photons must reach the detector (a CCD-equipped telescope,
248: a digital camera
249: or some other observation systems of the same kind).
250:
251: The next step in the simulation is to define the incidence direction of the decay
252: photon on the detector, situated on the Earth surface at the location $P_3$
253: (see Fig.~1). We generate uniformly the cosinus value of the ``local" zenith
254: angle $\theta_E$ inside the shadow cone (or in the cone subtended by the
255: telescope or by the analysis apperture if smaller)
256: and then the value of the ``local" azimuthal
257: angle $\phi_E$, as defined in Fig.~1.
258:
259: The probability of neutrino radiative decays (thus of their lifetime) depends on
260: their electric dipole momentum, which, for the weak flavor eigenstates, is
261: limited by the existing experimental data \cite{pdg}.
262: Those limits are not directly applicable for the mass eigenstates. We may assume that
263: the lifetime of $\nu_2$ is much larger than
264: the time of flight from the Sun to the Earth (otherwise the experiment would
265: be impossible). This implies that the decay points of the massive solar neutrinos
266: are uniformely distributed along their path from the Moon to the
267: Earth.
268:
269: As we already defined the
270: direction of the ``detected" photon, we cannot choose the decay point of the neutrino
271: using the above assumption; instead we take advantage of the nearly cylindrical
272: symmetry of our problem to choose
273: the decay point (point $P_3$ in Fig. 1)
274: randomly along the photon
275: path and then determine the neutrino path as the segment starting in $P_1$ and
276: ending in $P_3$. This does not affect the expected uniformity of the distribution
277: of the decay points along the neutrino path from the Moon to the Earth, as shown
278: in Fig.~3a. We also checked the isotropy on the incidence of
279: solar neutrinos on the Moon surface (see Fig. 3b).
280:
281:
282: \begin{figure}
283: \vspace{-17mm}
284: \begin{center}
285: \mbox{ \epsfysize=7cm
286: \epsffile{izo.eps}}
287: \caption{``Isotropy" tests for the simulated solar neutrino flux:
288: a) The distribution of neutrino path lengths from the Moon till their
289: decay point. b) The distribution of the points of incidence of the solar $\nu$'s
290: on the Moon disk.
291: One may notice that both distributions are uniform}
292: \end{center}
293: \end{figure}
294:
295:
296: \subsection{Neutrino decays}
297: \label{decay}
298:
299: For every simulated event we know (from the steps described
300: in the above subsections) the 4-vector of the ``heavy" neutrino $\nu_2$, its
301: birth and decay points, and the direction of the photon emission;
302: from the 4-momentum conservation we obtain the energy (in the
303: laboratory reference frame, LRF) of the emitted $\gamma$
304: \begin{equation}
305: E_\gamma = \frac{\Delta m^2}{2}~\frac{1}{E_\nu-p_\nu \cos\theta}~,
306: \end{equation}
307: where $E_\nu$ and $p_\nu$ are the energy and the momentum of the parent $\nu_2$
308: neutrino, respectively, and $\theta$ is the emission angle of the photon relative to
309: the direction of the $\nu_2$ momentum in the LRF (see Fig. 1).
310:
311: In the center of mass (CM) of the decaying $\nu_2$ neutrino, the probability
312: density of the emission of a photon at the zenithal angle $\theta^*$ is given by:
313: \begin{equation}
314: \frac{d\Gamma}{d(\cos \theta^*)}=K(1+\alpha \cos \theta^*)
315: \end{equation}
316: where the $\alpha$ parameter depends on the polarization state of the initial
317: $\nu_2$ flux: $\alpha = 0$ for unpolarized (Majorana) neutrinos, and
318: $\alpha = \mp 1$ for left and right handed (Dirac) $\nu_2$'s, respectively.
319: The constant $K$ comes from the general description of the decay of a fermion into
320: a fermion and a boson
321: \begin{equation}
322: K = \frac{\alpha_e^2}{\pi^2} \frac{m_2}{(\Delta m^2)^3}(m_1^2 + m_2^2 +
323: m_1 m_2),
324: \end{equation}
325: where $\alpha_e$ is the fine structure constant.
326:
327: In order to estimate the probability of each simulated event,
328: one would have to integrate Eq. 3 inside the solid angle under which the
329: emitted photon ``sees" the detector. This is unpractical since the physical
330: dimension of any usable detector is too small compared to the distance scales
331: involved in the simulation.
332: An equivalent approach is based on the observation that any
333: imaging system (such as a CCD in the focal plane of some optical system) has a limited
334: angular resolution, as the images of all point sources inside the solid angle covered
335: by a single pixel will be superimposed. We can then numerically integrate Eq. 3
336: on the area around the decay point $P_2$ (see Fig. 4 for a
337: sketch; the integration area is represented as the dark square centered in $P_2$)
338: that corresponds
339: to the angular aperture of a single pixel. The probabilities obtained
340: in this way are used as weights for the MC generated events.
341:
342:
343: \begin{figure}
344: \vspace{-10mm}
345: \begin{center}
346: \mbox{ \epsfysize=8cm
347: \epsffile{figmc.eps}}
348: \caption{A sketch showing the integration area for the probability density in Eq. 3.
349: Any decay event inside the dark square around point $P_2$ would lead to a signal
350: in the same CCD pixel. (In this sketch the Moon is not represented.)}
351: \end{center}
352: \end{figure}
353:
354:
355:
356: \section{Results and discussions}
357:
358:
359: The results presented in this Section are obtained assuming
360: an angular resolution of each pixel of 10", and an angular aperture
361: of the analysis (the maximum value of $\theta_E$ in Fig. 1) of 480",
362: as in the case of the digital
363: video-camera used during the 2001 TSE \cite{n3}. The mass $m_1$ of the $\nu_1$
364: mass eigenstate was considered in a range between $10^{-3}$ eV and 0.3 eV; MC
365: simulations
366: were made for $\Delta m^2 = 6 \times 10^{-5}$ eV$^2$ and also for $\Delta
367: m^2 = 2.5 \times
368: 10^{-3}$ eV$^2$. For the polarization parameter $\alpha$ in Eq. 3, we considered three
369: possible values: -1, 0 and 1.
370:
371: For each combination of neutrino mass and $\Delta m^2$ we
372: requested $3 \times 10^4$ unweighted events yielding photons in the visible range.
373: Assuming $\Delta m^2 = 6 \times 10^{-5}$ eV$^2$ the total number of iterations was
374: about $1.6 \times 10^9$, while for $\Delta m^2 = 2.5 \times 10^{-3}$ eV$^2$, about
375: $3.3 \times 10^7$ generated events were needed. The simulated events were then weighted
376: according to
377: the integral of Eq. 3, as discussed in the previous Section.
378: In weighting the events the factor $K$ in Eq. 3 was set to 1, as it is a constant
379: for each $(m_2,~\Delta m^2)$ hypothesis.
380: For this
381: analysis we also imposed the initialization of the random number
382: generator to be the same for each run.
383:
384: Fig. 5 shows the fraction of visible photons, versus
385: the mass of the lighter neutrino $\nu_1$. Fig. 5a corresponds to
386: $\Delta m^2 = 6 \times 10^{-5}$ eV$^2$, while the results for $\Delta m^2 = 2.5 \times
387: 10^{-3}$ eV$^2$ are presented in Fig. 5b. The light triangles are obtained
388: assuming $\alpha = 1$, while the dark and light circles correspond to $\alpha = 0$
389: and $\alpha = +1$, respectively.
390:
391:
392: \begin{figure}
393: \vspace{-65mm}
394: \begin{center}
395: \mbox{ \epsfysize=14cm
396: \epsffile{epsilon.eps}}
397: \caption{The fraction of visible photons produced in the simulated radiative decays
398: of massive solar neutrinos. The assumed squared mass differences are (a) $\Delta
399: m^2 = 6 \times
400: 10^{-5}$ eV$^2$ and (b) $2.5 \times 10^{-3}$ eV$^2$ (right). The light triangles,
401: light circles and dark circles correspond to polarisations $\alpha = -1$,
402: 0 and +1, respectively. The dashed lines are drawn only to guide the eye.}
403: \end{center}
404: \end{figure}
405:
406: As discussed in the previous Section, for each simulated neutrino decay event we
407: numericaly integrated Eq. 3, thus obtaining a probability that includes the
408: contributions from the kinematics of the decay itself as well as from the
409: {\em a priori} request included in the simulation that the emitted photon
410: reaches the detector. In Fig. 6 we present these probabilities, averaged over
411: all Monte Carlo events yielding visible photons,
412: versus the mass of the $\nu_1$ neutrino.
413: Fig. 6a refers to $\Delta m^2 = 6 \times 10^{-5}$ eV$^2$, and Fig. 6b
414: to $\Delta m^2 = 2.5 \times 10^{-3}$ eV$^2$. The symbols used for different
415: polarization states are the same as in Fig. 5.
416:
417:
418: \begin{figure}
419: \vspace{-65mm}
420: \begin{center}
421: \mbox{ \epsfysize=14cm
422: \epsffile{prob.eps}}
423: \caption{ Average probabilities for the neutrino radiative decay yielding
424: visible photons inside the simulated analysis acceptance.
425: The assumed squared mass differences are $\Delta
426: m^2 = 6 \times
427: 10^{-5}$ eV$^2$ (a) and $2.5 \times 10^{-3}$ eV$^2$ (b). The light triangles,
428: the light circles and the dark circles correspond to polarisations $\alpha = -1$,
429: 0 and +1, respectively. The dashed lines are only meant to guide the eye.}
430: \end{center}
431: \end{figure}
432:
433: The probabilities shown in Fig. 6 are very small, but one should consider that
434: they apply to all the solar massive neutrinos that cross the ``acceptance cone"
435: of the detector between the Earth and the Moon (see Fig. 1).
436:
437: Both Figs. 5 and 6 show a strongly non-linear behaviour, as the condition
438: imposed to the photons to be in the visible spectrum selects different regions
439: of the solar neutrino spectrum for different mass or polarization hypothesis. This
440: effect is illustrated in Fig. 7, for three values of the neutrino mass $m_1$,
441: 0.001, 0.01 and 0.1 eV: the solid, dashed and dotted histograms respectively.
442: Fig. 7a corresponds to $\Delta m^2 = 6 \times 10^{-5}$ eV$^2$, while
443: Fig. 7b to $\Delta m^2 = 2.5 \times 10^{-3}$ eV$^2$. In all cases the polarization
444: parameter $\alpha$ was assumed -1.
445:
446:
447: \begin{figure}
448: \vspace{-20mm}
449: \begin{center}
450: \mbox{ \epsfysize=8cm
451: \epsffile{fig7.eps}}
452: \caption{The energy distribution of the solar neutrinos that yield visible photons
453: through radiative decay, assuming $m_1 = 0.001$ eV (solid histograms), 0.01 eV
454: (dashed histograms) and 0.1 eV (dotted histograms). The squared mass difference
455: is assumed $6 \times 10^{-5}$ eV$^2$ (a) and $2.5 \times 10^{-3}$ eV$^2$ (b).
456: In all cases $\alpha = -1$.}
457: \end{center}
458: \end{figure}
459:
460: Fig. 7 suggests that the high energy solar $^8B$ neutrino tail does not
461: yield visible photons through radiative decays. Such process
462: could instead happen for low energy $pp$ neutrinos, with some contributions
463: from the $^{13}N$, $^{15}O$ neutrinos and from the $^7Be$ and $pep$ lines;
464: notice that these last contributions are more
465: noticeable in Fig. 7a than in Fig. 7b.
466:
467:
468: In conducting an experiment searching for visible photons from a hypothetical radiative
469: solar neutrino decay during TSE's and in choosing the appropriate data analysis
470: methodology, the simulation of the expected signal is very important. Fig. 8 presents
471: such simulations, in the same conditions and with the same conventions as in Fig. 7.
472: The shape of the signals corresponding to different neutrino masses and mass
473: differences (assuming left handed neutrinos, thus $\alpha = -1$) is presented as
474: the radial distribution of the ``detected" luminosity. This is equivalent to
475: the distribution of the weighted number of visible photons, averaged on
476: circular corona around the same value of the incidence zenith angle $\theta_E$
477: (see Fig. 1).
478:
479: \begin{figure}
480: %\vspace{-10mm}
481: \begin{center}
482: \mbox{ \epsfysize=8cm
483: \epsffile{fig8.eps}}
484: \caption{The expected shapes of the visible signals produced by the
485: hypothesized solar neutrino radiative decay,
486: assuming $m_1 = 0.001$ eV (solid histograms), 0.01 eV
487: (dashed) histograms) and 0.1 eV (dotted histograms). The squared mass difference
488: is assumed to be $6 \times 10^{-5}$ eV$^2$ (a) and $2.5 \times 10^{-3}$ eV$^2$ (b).
489: In all cases $\alpha = -1$.}
490: \end{center}
491: \end{figure}
492:
493:
494: The histograms in Fig. 8a correspond to the simulated $\nu_2 \rightarrow \nu_1 +
495: \gamma$ decays. For all neutrino masses, the expected signal is concentrated at
496: small $\theta_E$ angles (about 50 arcsec).
497: The widths and shapes of the signals are
498: sensitive to the mass assumed: the larger the mass, the narrower the signal band.
499: The two peaks seen at about 300 and 400 arcsec. for $m_1 = 10^{-3} eV$ could be
500: correlated with the contribution of $^7Be$ or $pep$ neutrinos; they are about
501: a factor 100 lower than the central maxima, so their contribution
502: is not measurable.
503:
504: In the case of $\nu_3 \rightarrow \nu_1 + \gamma$ simulated decays (Fig. 8b),
505: the signal is
506: broader (about 250 arcsec) and is less sensitive to the mass choice.
507: The peaks observed at about
508: 250" could have a similar origin as those in Fig. 8a, but could also be statistical
509: fluctuations.
510:
511: If an experiment
512: %looking for visible photons emitted by the radiative decay of
513: on solar neutrinos
514: has also a good energy resolution, then further information could
515: be obtained from the analysis of the spectra of the observed signal. Such spectra
516: are shown in Fig. 9, assuming only left-handed neutrinos, and considering the same
517: examples as in Figs. 7 and 8.
518:
519:
520: \begin{figure}
521: %\vspace{-10mm}
522: \begin{center}
523: \mbox{ \epsfysize=8cm
524: \epsffile{fig9.eps}}
525: \caption{The expected energy spectra of the visible signals produced by the
526: hypothesized solar neutrino radiative decay,
527: assuming $m_1 = 0.001$ eV (solid histograms), 0.01 eV
528: (dashed histograms) and 0.1 eV (dotted histograms). The squared mass difference
529: is assumed $6 \times 10^{-5}$ eV$^2$ (a) and $2.5 \times 10^{-3}$ eV$^2$ (b).
530: In all cases $\alpha = -1$.}
531: \end{center}
532: \end{figure}
533:
534: As for the shape of the signal, its spectral decomposition seems to be more
535: sensitive to the neutrino mass values for the $\nu_2 \rightarrow \nu_1 +
536: \gamma$ decays (Fig. 9a). In this case most of the visible photons
537: are ``detected" in the red part of the spectrum, while the spectra in Fig.
538: 9b suggest a dominant signal in the green, due to the larger mass difference
539: between $\nu_3$ and $\nu_1$.
540:
541: Let us assume that an experiment as those simulated in this work would measure,
542: during a TSE, an excess of visible photons $\Phi_{obs.}$ from the
543: direction of the center of the sun, or, if the
544: search yields a negative result,
545: $\Phi_{obs.}$ is the photon flux of the observed fluctuations. In the first case,
546: estimates of the neutrino lifetime could be done; otherwise, a lower experimental
547: limit could be deduced.
548: Assuming that solar (electron) neutrinos are superpositions of only two mass
549: eigenstates,
550: \begin{equation}
551: |\nu_e > = |\nu_1> \cos \theta + |\nu_2> \sin \theta,
552: \end{equation}
553: where $m_2 > m_1$ and $\theta$ is the mixing angle, the average lifetime (or its
554: lower limit) $\tau$ of the $\nu_2$ neutrino could be computed from
555: \begin{equation}
556: N_\gamma = P \Phi_2 S_M t_{obs} \left(1-e^{-\frac{<t_{ME}>}{\tau}} \right)
557: e^{-\frac{t_{SM}}{\tau}},
558: \end{equation}
559: where $N_\gamma$ is the number of decay visible photons observed,
560: %(or its fluctuation, if no signal revealed)
561: $P$ are the mass - dependent probabilities shown in
562: Fig. 6 a, $\Phi_2 = \Phi_\nu \sin^2 \theta$,(where $\Phi_\nu$ is the
563: flux of solar neutrinos at the Earth (or Moon) level), $S_M$ is the area of the
564: Moon surface covered by the analysis (the base of the cone of angle $\theta_E$
565: in Fig. 1) and $t_{obs}$ is the time of observation. $<t_{ME}>$ is the average
566: time spent by solar neutrinos inside the observation cone
567: (about one third of the flight
568: time from the Moon to the Earth), and $t_{SM}$ is the time of flight of the neutrinos
569: from the Sun to the Moon. The low numerical values of the probabilities $P$
570: are compensated by the large solar neutrino flux, combined with the large
571: area $S_M$, so an experiment as the one simulated here could yield at least
572: significant upper limits on the $\nu_2$ lifetime $\tau$.
573: In the conditions of a 3.5 minutes long TSE (as that of 2001) observed with an
574: instrument with characteristics similar to those considered in this simulation,
575: one would expect a $\nu_2$ lifetime (in the proper reference frame)
576: sensitivity ranging from few seconds to about $10^4$ seconds, assuming
577: $\nu_2$ neutrino masses of few $10^{-2}$ eV.
578:
579: \section{Conclusions}
580:
581: In this paper we presented a Monte Carlo simulation for an experiment looking
582: for visible photons emitted by
583: a possible solar neutrino radiative decay, during a total solar
584: eclipse. It was shown that for neutrino masses smaller than 0.1
585: eV and assuming squared mass differences in agreement with the Large Mixing
586: Angle Solution (LMA)
587: of the solar neutrino oscillations
588: \cite{sk}, such an experiment could give also an estimate
589: of the neutrino mass.
590:
591: The analysis of the experimental data collected by some of the authors during the
592: 2001 TSE in Zambia is being completed using the simulation results obtained
593: in this paper.
594:
595: \section{Acknowledgements}
596:
597: We would like to aknowledge many colleagues for useful comments and discussions.
598:
599: This work was founded by NATO Grant PST.CLG.977691 and partially supported by the
600: Italian Space Agency (ASI) and the Romanian Space Agency (ROSA).
601:
602:
603:
604:
605: \begin{thebibliography}{00}
606:
607: % \bibitem{label}
608: % Text of bibliographic item
609:
610: % notes:
611: % \bibitem{label} \note
612:
613: % subbibitems:
614: % \begin{subbibitems}{label}
615: % \bibitem{label1}
616: % \bibitem{label2}
617: % If there is a note, it should come last:
618: % \bibitem{label3} \note
619: % \end{subbibitems}
620:
621: \bibitem{gg} G. Giacomelli and M. Sioli, Astroparticle Physics, hep-ex/0211035
622: \bibitem{sciama} A.L. Melott, D.W. Sciama,
623: {\em Phys.Rev.Lett.} {\bf 46} (1981) 1369-1372.
624: \bibitem{sciama2} D.W. Sciama, {\em Nucl.Phys.Proc.Suppl.} {\bf38} (1995) 320-323.
625: \bibitem{vanucci} C. Birnbaun et al., {\em Phys. Lett.} {\bf B397} (1997) 143-146.
626: \bibitem{frere} J.-M. Fr\`{e}re and D. Monderen, {\em Phys. Lett.} {\bf B431} (1998) 368-373.
627: \bibitem{n1} S. Cecchini et al. (NOTTE Coll.),
628: {\em Astrophys. and Space Sci.} {\bf 273} (2000) 35-41.
629: \bibitem{n2} S. Cecchini et al. (NOTTE Coll.),
630: Limits on radiative decays of solar neutrinos
631: from a measurement during a solar eclipse, hep-ex/0011048
632: \bibitem{n3} V. Popa et al. (NOTTE Coll.),
633: {\em Astrophys. and Space
634: Sci.} {\bf 282} (2002) 235-244.
635: \bibitem{bahcall1} J.N. Bahcall, Standard Solar Models, astro-ph/9808162
636: \bibitem{sno1} Q.R. Ahmad et al. (SNO Coll.), {\em Phys. Rev. Lett.} {\bf 87}
637: (2001) 071301.
638: \bibitem{sno2} Q.R. Ahmad et al. (SNO Coll.), {\em Phys. Rev. Lett.} {\bf 89}
639: (2002) 011301.
640: \bibitem{map} D.N. Spergel at al., First year Wilkinson Microwave Anisotropy
641: Probe (WMAP) observations: Determination of cosmological parameters,
642: {\em Ap. J.} (2003) -accepted-
643: \bibitem{macro} M. Ambrosio et al. (MACRO Coll.),
644: {\em Phys. Lett.} {\bf B434} (1998) 451; {\em Phys. Lett.} {\bf B357} (1995) 481.
645: \bibitem{miri} M. Ambrosio et al. (MACRO Coll.),
646: {\em Phys. Lett.} {\bf B566} (2003) 35.
647: \bibitem{sk} Y. Fukuda et al. (Super-Kamiokande Coll.), {\em Phys. Rev. Lett.}
648: {\bf 81} (1998) 1562; {\em Nucl. Instrum. Meth.} {\bf 503} (2003) 114.
649: \bibitem{bahcall2} J.N. Bahcall, M.H. Pinsonneault and S. Basu, {\em Ap. J.}
650: {\bf 555} (2001) 990-1012.
651: \bibitem{web} http://www.sns.ias.edu/\~{} jnb/
652: \bibitem{pdg} K. Hagiwara et al. (Particle Data Group), {\em Phys. Rev. } {\bf
653: D66} (2002) 010001.
654:
655: \end{thebibliography}
656:
657:
658: \end{document}
659:
660: