1: \documentclass[twoside]{article}
2: \usepackage{fleqn,espcrc2}
3: \usepackage{graphicx}
4: \usepackage{amssymb}
5: \usepackage{float}
6: %\bibliographystyle{physrev3}
7: %\usepackage[figuresright]{rotating}
8:
9: % put your own definitions here:
10: \newcommand{\ttbs}{\char'134}
11: \newcommand{\AmS}{{\protect\the\textfont2
12: A\kern-.1667em\lower.5ex\hbox{M}\kern-.125emS}}
13:
14: % add words to TeX's hyphenation exception list
15: \hyphenation{author another created financial paper re-commend-ed Post-Script}
16:
17: % declarations for front matter
18: \title{Four-Neutrino Scenarios\thanks{
19: Talk presented at
20: NOW 2000, Conca Specchiulla (Otranto, Italy), 9-16 Sep. 2000;
21: DFTT 47/00, hep-ph/0012236.}
22: }
23:
24: \author{C. Giunti\thanks{
25: I would like to thank G. Mills and B. Louis
26: for useful information
27: on the LSND experiment.
28: }\\[0.2cm]
29: INFN, Sez. di Torino, and Dip. di Fisica Teorica,
30: Univ. di Torino, I--10125 Torino, Italy}
31:
32: \begin{document}
33:
34: \begin{abstract}
35: The main features of four-neutrino 3+1 and 2+2 mixing schemes
36: are reviewed,
37: after a discussion on the necessity of
38: at least four massive neutrinos
39: if the solar, atmospheric and LSND anomalies are
40: due to neutrino oscillations.
41: \end{abstract}
42:
43: \maketitle
44:
45: \begin{figure*}[t!]
46: \begin{center}
47: \includegraphics[bb=13 720 522 825,width=0.80\textwidth]{4schemes.eps}
48: \end{center}
49: \caption{ \label{4schemes}
50: Qualitative illustration of the possible four-neutrino schemes.
51: }
52: \end{figure*}
53:
54: \section{Introduction}
55: \label{Introduction}
56:
57: Solar and atmospheric neutrino experiments
58: have observed for a long time
59: anomalies that are commonly interpreted
60: as evidences in favor of neutrino oscillations
61: with mass squared differences
62: \begin{eqnarray}
63: &&
64: 10^{-11} \, \mathrm{eV}^2
65: \lesssim
66: \Delta{m}^2_{\mathrm{SUN}}
67: \lesssim
68: 10^{-4} \, \mathrm{eV}^2
69: \,,
70: \label{dm2sun}
71: \\
72: &&
73: 10^{-3} \, \mathrm{eV}^2
74: \lesssim
75: \Delta{m}^2_{\mathrm{ATM}}
76: \lesssim
77: 10^{-2} \, \mathrm{eV}^2
78: \,,
79: \label{dm2atm}
80: \end{eqnarray}
81: respectively
82: (see Refs.\cite{NOW2000-Smirnov,NOW2000-Kajita}).
83: More recently,
84: the accelerator LSND experiment has reported
85: the observation of
86: $\bar\nu_\mu\to\bar\nu_e$
87: and
88: $\nu_\mu\to\nu_e$
89: appearance~\cite{NOW2000-Spentzouris}
90: with a mass-squared difference
91: \begin{equation}
92: 10^{-1} \, \mathrm{eV}^2
93: \lesssim
94: \Delta{m}^2_{\mathrm{LSND}}
95: \lesssim
96: 10 \, \mathrm{eV}^2
97: \,.
98: \label{dm2LSND}
99: \end{equation}
100: The LSND evidence in favor of neutrino oscillations has not been confirmed
101: by other experiments,
102: but it has not been excluded either.
103: Awaiting an independent check of the LSND result,
104: that will probably come soon
105: from the MiniBooNE experiment~\cite{NOW2000-Spentzouris},
106: it is interesting to consider the possibility
107: that the results of solar, atmospheric and LSND experiments
108: are due to neutrino oscillations.
109: In this case,
110: the existence of the three mass-squared differences
111: (\ref{dm2sun})--(\ref{dm2LSND})
112: with different scales
113: implies that there are at least four massive neutrinos
114: (three massive neutrinos are not enough because
115: the three $\Delta{m}^2$'s
116: have different scales and
117: do not add up to zero).
118:
119: Since the mass-squared differences
120: (\ref{dm2sun})--(\ref{dm2LSND})
121: have been obtained by analyzing separately the data
122: of each type of experiment
123: (solar, atmospheric and LSND)
124: in terms of two-neutrino mixing,
125: it is legitimate to ask if
126: three different mass squared are really necessary to fit the data.
127: The answer is ``yes'',
128: as explained in Section~\ref{Three}.
129:
130: Although the precise measurement of the invisible width
131: of the $Z$ boson has determined that there are only three
132: active flavor neutrinos,
133: $\nu_e$, $\nu_\mu$, $\nu_\tau$,
134: the possible existence of at least four massive neutrinos is not a problem,
135: because in general flavor neutrinos are not mass eigenstates,
136: \textit{i.e.} there is \emph{neutrino mixing}
137: (see, \textit{e.g.}, Ref.\cite{BGG-review-98-brief}).
138:
139: In general,
140: the left-handed component
141: $\nu_{\alpha L}$
142: of a flavor neutrino field
143: is a linear combination of the left-handed components
144: $\nu_{kL}$
145: of neutrino fields with masses $m_k$:
146: %\begin{equation}
147: $
148: \nu_{\alpha L}
149: =
150: \sum_{k} U_{\alpha k} \nu_{kL}
151: $,
152: %\,,
153: %\label{mixing}
154: %\end{equation}
155: where $U$ is the unitary neutrino mixing matrix.
156: The number of massive neutrinos is only constrained
157: to be $\geq3$.
158: Following the old principle known as \emph{Occam razor},
159: we consider the simplest case of four massive neutrinos
160: that allows to explain all data
161: with neutrino oscillations~\cite{www-4nu}.
162: In this case,
163: in the flavor basis the usual three active neutrinos
164: $\nu_e$, $\nu_\mu$, $\nu_\tau$,
165: are associated with a sterile neutrino,
166: $\nu_s$,
167: that is a singlet of the electroweak group.
168:
169: Taking into account the measured hierarchy
170: \begin{equation}
171: \Delta{m}^2_{\mathrm{SUN}}
172: \ll
173: \Delta{m}^2_{\mathrm{ATM}}
174: \ll
175: \Delta{m}^2_{\mathrm{LSND}}
176: \,,
177: \label{hierarchy}
178: \end{equation}
179: %(from Eqs.(\ref{dm2sun})--(\ref{dm2LSND})),
180: there are only six types of possible four-neutrino schemes,
181: which are shown in Fig.\ref{4schemes}.
182: These six schemes are divided in two classes:
183: 3+1 and 2+2.
184: In both classes
185: there are two groups of neutrino masses
186: separated by the LSND gap, of the order of 1 eV,
187: such that the largest
188: mass-squared difference
189: generates the oscillations observed in the LSND experiment:
190: $\Delta{m}^2_{\mathrm{LSND}} = |\Delta{m}^2_{41}|$
191: (where
192: $\Delta{m}^2_{kj} \equiv m_k^2 - m_j^2$).
193: In 3+1
194: schemes there is a group of three neutrino masses
195: separated from an isolated mass
196: by the LSND gap.
197: In 2+2 schemes
198: there are two pairs of close masses
199: separated by the LSND gap.
200: The numbering of the mass eigenvalues in Fig.~\ref{4schemes}
201: is conveniently chosen in order to have always
202: solar neutrino oscillations generated by
203: $\Delta{m}^2_{21} = \Delta{m}^2_{\mathrm{SUN}}$.
204: In 3+1 schemes
205: atmospheric neutrino oscillations are generated by
206: $
207: |\Delta{m}^2_{31}|
208: \simeq
209: |\Delta{m}^2_{32}|
210: =
211: \Delta{m}^2_{\mathrm{ATM}}
212: $,
213: whereas in 2+2 schemes
214: they are generated by
215: $
216: |\Delta{m}^2_{43}|
217: =
218: \Delta{m}^2_{\mathrm{ATM}}
219: $.
220:
221: In 1999
222: the 3+1 schemes were rather strongly disfavored by the experimental data,
223: with respect to the 2+2 schemes~\cite{BGGS-AB-99-brief}.
224: %\cite{BGG-AB-96,Barger-variations-98,BGGS-AB-99,Giunti-JHU-99}.
225: In June 2000 the LSND collaboration presented the
226: results of a new improved analysis of their data,
227: leading to an allowed region in the
228: $\sin^2 2\vartheta$--$\Delta{m}^2$ plane
229: ($\vartheta$ is the two-generation mixing angle)
230: that is larger and shifted towards lower values of
231: $\sin^2 2\vartheta$,
232: with respect to the 1999 allowed region.
233: This implies that the 3+1 schemes are now marginally
234: compatible with the data.
235: Therefore,
236: in Section~\ref{3+1}
237: I discuss the 3+1 schemes,
238: that have been recently revived~\cite{Barger-Fate-2000-brief,%
239: Giunti-Laveder-ata-00,%
240: Peres-Smirnov-3+1-00}.
241: In Section~\ref{2+2}
242: I discuss the 2+2 schemes,
243: that are still favored by the data.
244:
245: \section{Three $\Delta{m}^2$'s are necessary}
246: \label{Three}
247:
248: Let us consider the general expression
249: of the probability of
250: $\nu_\alpha\to\nu_\beta$
251: transitions in vacuum valid for any number
252: of massive neutrinos:
253: \begin{equation}
254: P_{\nu_\alpha\to\nu_\beta}
255: =
256: \left|
257: \sum_k
258: U_{\alpha k}^* U_{\beta k}
259: \exp\left(-i\frac{\Delta{m}^2_{kj}L}{2E}\right)
260: \right|^2
261: \,,
262: \label{prob}
263: \end{equation}
264: where
265: $L$ is the source-detector distance,
266: $E$ is the neutrino energy,
267: and
268: $j$ is anyone of the mass eigenstate indices
269: (a phase common to all terms in the sum in Eq.(\ref{prob}) is irrelevant).
270:
271: If all the phases
272: $\Delta{m}^2_{kj}L/2E$'s
273: are very small,
274: oscillations are not observable because
275: the probability reduces to
276: $P_{\nu_\alpha\to\nu_\beta} \simeq \delta_{\alpha\beta}$.
277: Since the LSND experiment has the smallest
278: average $L/E$,
279: of the order of $1 \, \mathrm{eV}^{-2}$,
280: at least one $\Delta{m}^2_{kj}$,
281: denoted by
282: $\Delta{m}^2_{\mathrm{LSND}}$,
283: must be larger than about
284: $10^{-1} \, \mathrm{eV}^2$
285: in order to generate the observed
286: $\bar\nu_\mu\to\bar\nu_e$
287: and
288: $\nu_\mu\to\nu_e$
289: LSND transitions,
290: whose measured probability
291: is of the order of $10^{-3}$.
292:
293: Solar neutrino experiments observe
294: large transitions of $\nu_e$'s into other states,
295: with an average probability of about 1/2.
296: These transitions
297: cannot be generated by a
298: $\Delta{m}^2_{kj} \gtrsim 10^{-3} \, \mathrm{eV}^2$
299: because they should have been observed by the
300: long-baseline CHOOZ experiment~\cite{CHOOZ-99-brief}.
301: Hence,
302: at least another $\Delta{m}^2_{kj}$
303: smaller than about
304: $10^{-3} \, \mathrm{eV}^2$,
305: denoted by
306: $\Delta{m}^2_{\mathrm{SUN}}$,
307: is needed for the oscillations of solar neutrinos.
308:
309: The necessary existence of at least a third $\Delta{m}^2_{kj}$
310: for atmospheric neutrino oscillations
311: is less obvious,
312: but can be understood by noticing that
313: a dependence of the transition probability
314: from the energy $E$
315: and/or from the distance $L$
316: is observable only if at least one phase
317: $\Delta{m}^2_{kj}L/2E$
318: is of order one.
319: Indeed, all the exponentials with phase
320: $\Delta{m}^2_{kj}L/2E \ll 1$
321: can be approximated to one,
322: whereas
323: all the exponentials with phase
324: $\Delta{m}^2_{kj}L/2E \gg 1$
325: are washed out by the averages
326: over the energy resolution of the detector
327: and the uncertainty in the source-detector distance.
328: Since the Super-Kamiokande atmospheric neutrino experiment
329: measures a variation of the oscillation probability
330: for
331: $L/E \sim 10^2 \div 10^3 \, \mathrm{eV}^{-2}$
332: (see Ref.\cite{NOW2000-Kajita}),
333: there must be at least one $\Delta{m}^2_{kj}$
334: in the range
335: $10^{-3} \div 10^{-2} \, \mathrm{eV}^2$,
336: which is out of the ranges allowed for
337: $\Delta{m}^2_{\mathrm{SUN}}$
338: and
339: $\Delta{m}^2_{\mathrm{LSND}}$.
340: Therefore,
341: at least a third $\Delta{m}^2_{kj}$,
342: denoted by
343: $\Delta{m}^2_{\mathrm{ATM}}$,
344: is needed for atmospheric neutrino oscillations.
345: This argument is supported by a detailed calculation
346: presented in Ref.\cite{Fogli-Lisi-Marrone-Scioscia-no3-99}.
347:
348: \begin{figure}[b!]
349: \begin{center}
350: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{uel4.eps}
351: \end{center}
352: \caption{ \label{uel4}
353: 3+1 schemes.
354: Dotted and dashed lines:
355: $|U_{e4}|^2_{\mathrm{max}}$
356: from Bugey~\protect\cite{Bugey-brief}
357: and
358: CHOOZ~\protect\cite{CHOOZ-99-brief}.
359: Solid lines enclose the allowed regions.
360: }
361: \end{figure}
362:
363: In the following sections we discuss some phenomenological aspects
364: of the four-neutrino schemes in Fig.\ref{4schemes},
365: in which there are three mass squared differences
366: with the hierarchy (\ref{hierarchy})
367: indicated by the data.
368:
369: \begin{figure}[b!]
370: \begin{center}
371: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{umu4.eps}
372: \end{center}
373: \caption{ \label{umu4}
374: 3+1 schemes.
375: Dotted and dashed lines:
376: $|U_{\mu4}|^2_{\mathrm{max}}$
377: from CDHS~\protect\cite{CDHS-brief}
378: and
379: Super-Kamiokande~\protect\cite{NOW2000-Kajita,BGGS-AB-99-brief}.
380: Solid lines:
381: allowed regions.
382: }
383: \end{figure}
384:
385: \section{3+1 Schemes}
386: \label{3+1}
387:
388: In 3+1 schemes
389: the amplitude of
390: $\nu_\alpha\to\nu_\beta$ and $\nu_\beta\to\nu_\alpha$
391: transitions in short-baseline neutrino oscillation experiments
392: (equivalent to the usual
393: $\sin^2 2\vartheta$
394: in the two-generation case)
395: is given by
396: (see, for example, Ref.\cite{BGG-review-98-brief})
397: \begin{equation}
398: A_{\alpha\beta}
399: =
400: A_{\beta\alpha}
401: =
402: 4 |U_{\alpha4}|^2 |U_{\beta4}|^2
403: \,,
404: \label{14}
405: \end{equation}
406: and the oscillation amplitude
407: (again equivalent to the usual two-generation $\sin^2 2\vartheta$)
408: in short-baseline $\nu_\alpha$ disappearance experiments
409: is given by
410: \begin{equation}
411: B_\alpha
412: =
413: \sum_{\beta\neq\alpha} A_{\alpha\beta}
414: =
415: 4 \, |U_{\alpha4}|^2 \left( 1 - |U_{\alpha4}|^2 \right)
416: \,.
417: \label{16}
418: \end{equation}
419:
420: \begin{figure}[b!]
421: \begin{center}
422: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{31-amuel-allowed-slide.eps}
423: \end{center}
424: \caption{ \label{31-amuel-allowed}
425: 3+1 schemes.
426: Very thick solid line:
427: allowed regions.
428: Thick solid line:
429: disappearance bound (\ref{31bound}).
430: Dotted line:
431: LSND 2000 allowed regions at 90\% CL~\protect\cite{NOW2000-Spentzouris}.
432: Solid line:
433: LSND 2000 allowed regions at 99\% CL~\protect\cite{NOW2000-Spentzouris}.
434: Broken dash-dotted line:
435: Bugey exclusion curve at 90\% CL~\protect\cite{Bugey-brief}.
436: Vertical dash-dotted line:
437: CHOOZ exclusion curve at 90\% CL~\protect\cite{CHOOZ-99-brief}.
438: Long-dashed line:
439: KARMEN 2000 exclusion curve at 90\% CL~\protect\cite{NOW2000-Steidl}.
440: Short-dashed line:
441: BNL-E776 exclusion curve at 90\% CL~\protect\cite{BNL-E776}.
442: }
443: \end{figure}
444:
445: \begin{figure}[b!]
446: \begin{center}
447: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{amuta.eps}
448: \end{center}
449: \caption{ \label{amuta}
450: 3+1 schemes with $|U_{s4}|^2 \ll 1$.
451: Solid lines enclose the allowed regions.
452: Long dashed line:
453: CHORUS exclusion curve at 90\% CL~\protect\cite{NOW2000-Zucchelli}.
454: Short dashed line:
455: NOMAD exclusion curve at 90\% CL~\protect\cite{NOW2000-Vidal-Sitjes}.
456: Dotted line:
457: CDHS exclusion curve at 90\% CL~\protect\cite{CDHS-brief}.
458: }
459: \end{figure}
460:
461: Short-baseline $\bar\nu_e$ and $\nu_\mu$
462: disappearance experiments
463: put rather stringent limits
464: $B_e \leq B_e^{\mathrm{max}}$
465: and
466: $B_\mu \leq B_\mu^{\mathrm{max}}$
467: for
468: $|\Delta{m}^2_{41}|$
469: in the LSND-allowed region.
470: Taking into account also
471: the results of solar and atmospheric neutrino experiments,
472: Eq.(\ref{16}) implies that
473: $|U_{e4}|^2$
474: and
475: $|U_{\mu4}|^2$
476: are small
477: (see Ref.\cite{Giunti-Laveder-ata-00} and references therein):
478: \begin{equation}
479: |U_{e4}|^2 \leq |U_{e4}|^2_{\mathrm{max}}
480: \quad \mbox{and} \quad
481: |U_{\mu4}|^2 \leq |U_{\mu4}|^2_{\mathrm{max}}
482: \,,
483: \label{24}
484: \end{equation}
485: as shown by the dashed and dotted lines
486: in Figs.\ref{uel4} and \ref{umu4}.
487: These limits imply that
488: the amplitude
489: $A_{\mu e}$,
490: equivalent to the usual $\sin^2 2\vartheta$
491: in short-baseline
492: $\nu_\mu\to\nu_e$
493: and
494: $\bar\nu_\mu\to\bar\nu_e$
495: experiments,
496: is very small:
497: \begin{equation}
498: A_{{\mu}e}
499: \leq
500: 4 |U_{\mu4}|^2_{\mathrm{max}} |U_{e4}|^2_{\mathrm{max}}
501: \,,
502: \label{31bound}
503: \end{equation}
504: so small to be at the border of compatibility
505: with the oscillations
506: observed in the LSND experiment.
507: Figure~\ref{31-amuel-allowed}
508: shows the comparison of the bound (\ref{31bound})
509: with
510: the LSND allowed region,
511: taking into account also the exclusion curves
512: exclusion curves of the KARMEN~\cite{NOW2000-Steidl}
513: and BNL-E776~\cite{BNL-E776} experiments.
514: One can see that there are four regions
515: that are marginally allowed,
516: denoted by R1, R2, R3, R4.
517:
518: Let us denote by
519: $A_{{\mu}e}^{\mathrm{min}}$
520: the lower limit for $A_{{\mu}e}$
521: in the four allowed regions
522: in Fig.\ref{31-amuel-allowed}.
523: Then,
524: from
525: $A_{{\mu}e} = 4 |U_{\mu4}|^2 |U_{e4}|^2$
526: and the upper bounds (\ref{24}),
527: one can derive lower limits for
528: $|U_{e4}|^2$
529: and
530: $|U_{\mu4}|^2$:
531: \begin{equation}
532: |U_{e4}|^2
533: \geq
534: \frac{A_{{\mu}e}^{\mathrm{min}}}{4 |U_{\mu4}|^2_{\mathrm{max}}}
535: \,,
536: \quad
537: |U_{\mu4}|^2
538: \geq
539: \frac{A_{{\mu}e}^{\mathrm{min}}}{4 |U_{e4}|^2_{\mathrm{max}}}
540: \,.
541: \label{124}
542: \end{equation}
543: The upper and lower limits (\ref{24}) and (\ref{124})
544: for
545: $|U_{e4}|^2$ and $|U_{\mu4}|^2$
546: determine the allowed regions enclosed by solid lines
547: in Figs.\ref{uel4} and \ref{umu4}.
548:
549: Summarizing the general properties of 3+1 schemes obtained so far,
550: from Fig.\ref{uel4}
551: we know that $|U_{e4}|^2$ is very small,
552: of the order of $10^{-2}$,
553: and from Fig.\ref{umu4}
554: we know that in the regions R2, R3, R4
555: $|U_{\mu4}|^2$ is also very small,
556: of the order of $10^{-2}$,
557: whereas in the region R1
558: $|U_{\mu4}|^2$
559: is relatively large,
560: $0.33 \lesssim |U_{\mu4}|^2 \lesssim 0.55$.
561: On the other hand,
562: the mixings of $\nu_\tau$ and $\nu_s$ with $\nu_4$
563: are unknown.
564:
565: The authors of Ref.\cite{Barger-Fate-2000-brief}
566: considered the interesting possibility that
567: \begin{equation}
568: 1 - |U_{s4}|^2 \ll 1
569: \,,
570: \label{fate}
571: \end{equation}
572: \textit{i.e.}
573: that the isolated neutrino $\nu_4$
574: practically coincides with $\nu_s$.
575: Notice, however,
576: that $|U_{s4}|^2$
577: cannot be exactly equal to one,
578: because LSND oscillations require that
579: $|U_{e4}|^2$ and $|U_{\mu4}|^2$
580: do not vanish,
581: as shown in Figs.\ref{uel4} and \ref{umu4},
582: and unitarity implies that
583: $1 - |U_{s4}|^2 \geq |U_{e4}|^2 + |U_{\mu4}|^2$.
584: The possibility (\ref{fate})
585: is attractive because
586: it represents a perturbation of the standard
587: three-neutrino mixing
588: in which a mass eigenstate is added,
589: that mixes mainly with the new sterile neutrino $\nu_s$
590: and very weakly with
591: the standard active neutrinos $\nu_e$, $\nu_\mu$, $\nu_\tau$.
592: In this case,
593: the usual phenomenology of three-neutrino mixing
594: in solar and atmospheric
595: neutrino oscillation experiments is practically unchanged:
596: the atmospheric neutrino anomaly would be explained by
597: dominant $\nu_\mu\to\nu_\tau$ transitions,
598: with possible sub-dominant $\nu_\mu\leftrightarrows\nu_e$
599: transitions constrained by the CHOOZ bound,
600: and
601: the solar neutrino problem would be explained by an
602: approximately equal mixture of
603: $\nu_e\to\nu_\mu$ and $\nu_e\to\nu_\tau$ transitions
604: (see, for example, Ref.\cite{BGG-review-98-brief}).
605: An appealing characteristic of this scenario
606: is the practical absence of transitions
607: of solar and atmospheric neutrinos into
608: sterile neutrinos,
609: that seems to be favored
610: by the latest data
611: (see \cite{NOW2000-Suzuki,NOW2000-Kajita,NOW2000-Ronga}).
612:
613: \begin{figure}[b!]
614: \begin{center}
615: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{22-amuel-allowed-slide.eps}
616: \end{center}
617: \caption{ \label{22-amuel-allowed}
618: 2+2 schemes.
619: See caption of Fig.\ref{31-amuel-allowed}.
620: }
621: \end{figure}
622:
623: \begin{figure}[b!]
624: \begin{center}
625: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{sel.eps}
626: \end{center}
627: \caption{ \label{sel}
628: 2+2 schemes.
629: Dotted and dashed lines:
630: upper limit
631: from Bugey~\protect\cite{Bugey-brief}
632: and
633: CHOOZ~\protect\cite{CHOOZ-99-brief}.
634: The regions marked by ``A'' enclosed by solid lines are
635: allowed.
636: }
637: \end{figure}
638:
639: \begin{figure}[b!]
640: \begin{center}
641: \includegraphics[bb=89 420 540 730,width=0.45\textwidth]{smu.eps}
642: \end{center}
643: \caption{ \label{smu}
644: 2+2 schemes.
645: Solid, dotted and dashed lines:
646: limits
647: from CDHS~\protect\cite{CDHS-brief},
648: Super-Kamiokande~\protect\cite{NOW2000-Kajita,BGGS-AB-99-brief}
649: and
650: LSND~\protect\cite{NOW2000-Spentzouris,Giunti-JHEP-00-brief}.
651: The allowed regions are marked by ``A''.
652: }
653: \end{figure}
654:
655: Another interesting possibility has been considered
656: in Ref.\cite{Giunti-Laveder-ata-00}:
657: \begin{equation}
658: |U_{s4}|^2 \ll 1
659: \,.
660: \label{large}
661: \end{equation}
662: This could be obtained,
663: for example,
664: in the hierarchical scheme I (see Fig.~\ref{4schemes})
665: with an appropriate symmetry keeping the
666: sterile neutrino very light,
667: \textit{i.e.}
668: mostly mixed with the lightest mass eigenstates.
669: Notice that nothing forbids
670: $|U_{s4}|^2$
671: to be even zero exactly.
672: The possibility (\ref{large})
673: is interesting because if it is realized there are relatively large
674: $\nu_\mu\to\nu_\tau$ and $\nu_e\to\nu_\tau$ transitions
675: in short-baseline neutrino oscillation experiments,
676: that could be observed in the near future.
677: This is due to the fact that
678: the unitarity of the mixing matrix implies that
679: $|U_{\tau4}|^2$
680: is large
681: ($1-|U_{\tau4}|^2\ll1$ in the regions R2, R3, R4
682: and
683: $0.45 \lesssim |U_{\tau4}|^2 \lesssim 0.67$ in the region R1).
684: Therefore,
685: the amplitudes
686: $A_{\mu\tau} = 4 |U_{\mu4}|^2 |U_{\tau4}|^2$
687: and
688: $A_{e\tau} = 4 |U_{e4}|^2 |U_{\tau4}|^2$
689: of short-baseline
690: $\nu_\mu\to\nu_\tau$ and $\nu_e\to\nu_\tau$
691: oscillations
692: are suppressed only by the smallness of
693: $|U_{\mu4}|^2$ and $|U_{e4}|^2$
694: and lie just below the upper limits imposed by
695: the negative results of short-baseline
696: $\nu_\mu$ and $\bar\nu_e$
697: disappearance experiments.
698: Figure \ref{amuta}
699: shows the allowed regions
700: in the
701: $A_{\mu\tau}$--$|\Delta{m}^2_{41}|$
702: plane.
703: One can see that the region R4
704: is excluded by the negative results
705: of the CHORUS \cite{NOW2000-Zucchelli}
706: and NOMAD \cite{NOW2000-Vidal-Sitjes} experiments.
707: The other three regions are possible
708: and predict relatively large
709: oscillation amplitudes that could be observed
710: in the near future,
711: especially the two regions R2 and R3
712: in which
713: $A_{\mu\tau} \sim 4 \times 10^{-2} - 10^{-1}$.
714: An unattractive feature of this scenario
715: is its predictions of large $\nu_\mu\to\nu_s$ transitions
716: of atmospheric neutrinos,
717: that appear to be disfavored
718: by the latest data
719: (see \cite{NOW2000-Kajita,NOW2000-Ronga}).
720:
721: \null \vspace{-1cm} \null
722:
723: \section{2+2 Schemes}
724: \label{2+2}
725:
726: The two 2+2 schemes in Fig.~\ref{4schemes}
727: are favored by the data because they do not suffer the
728: constraint imposed by the thick solid line in Fig.\ref{31-amuel-allowed},
729: that is valid only in 3+1 schemes.
730: Therefore,
731: all the part of the LSND region
732: in the $A_{\mu e}$--$\Delta{m}^2_{41}$ plane
733: that is not excluded
734: by other experiments is allowed,
735: as shown in Fig.\ref{22-amuel-allowed}.
736: For this reason,
737: the phenomenology of 2+2 schemes has been studied
738: in many articles~\cite{www-4nu}.
739:
740: Figures~\ref{sel} and \ref{smu}
741: show the limits on the mixing of $\nu_e$ and $\nu_\mu$
742: obtained from the results of short-baseline,
743: solar and atmospheric experiments
744: \cite{www-4nu,Giunti-JHEP-00-brief}.
745: From Fig.~\ref{sel}
746: one can see that the mixing of $\nu_e$
747: with $\nu_3$ and $\nu_4$,
748: whose mass-squared difference
749: $\Delta{m}^2_{43}$
750: generates
751: atmospheric neutrino transitions,
752: is very small,
753: leading to a suppression of oscillations of $\nu_e$'s
754: in atmospheric and long-baseline experiments
755: \cite{BGG-bounds-98-brief}.
756:
757: The mixing of $\nu_\tau$ and $\nu_s$
758: is almost unknown,
759: with weak limits obtained in recent fits
760: of solar
761: \cite{Concha-SNO-00}
762: and atmospheric data
763: \cite{Yasuda-fouratm-00,Fogli-Lisi-Marrone-fouratm-00-brief}.
764: For example,
765: it is possible that
766: both solar $\nu_e$'s and atmospheric $\nu_\mu$'s
767: oscillate into approximately equal mixtures of
768: $\nu_\tau$ and $\nu_s$'s.
769:
770: In the future it may be possible to exclude
771: the scheme A
772: if it will be established with confidence
773: that the effective number of neutrinos
774: in Big-Bang Nucleosynthesis
775: is less that four.
776: In this case
777: $|U_{s3}|^2+|U_{s4}|^2 \ll 1$
778: \cite{Okada-Yasuda-97-brief}
779: and solar and atmospheric neutrino oscillations
780: occur, respectively,
781: through the decoupled channels
782: $\nu_e\to\nu_s$
783: and
784: $\nu_\mu\to\nu_\tau$.
785: It has been shown that in this scenario
786: the small mass splitting in scheme A between $\nu_3$ and $\nu_4$
787: is incompatible with radiative corrections
788: \cite{Ibarra-Navarro-4nu-00-brief}
789: and the effective Majorana mass in neutrinoless double-beta decay
790: in scheme A
791: is at the border of compatibility
792: with the experimental limit
793: \cite{Giunti-Neutrinoless-99-brief}.
794:
795:
796: \section{Conclusions}
797: \label{Conclusions}
798:
799: Four-neutrino mixing is a realistic possibility
800: (if the solar, atmospheric and LSND anomalies are
801: due to neutrino oscillations).
802: It is rather complicated, but very interesting,
803: both for theory and experiments,
804: because:
805: it has a rich phenomenology;
806: the existence of a sterile neutrino is far
807: beyond the Standard Model,
808: hinting for exciting new physics;
809: there are several observable oscillation channels
810: in short-baseline and long-baseline experiments;
811: CP violation may be observable in
812: long-baseline experiments
813: \cite{www-4nu}.
814:
815: %\bibliography{4nu}
816:
817: \begin{thebibliography}{10}
818:
819: \bibitem{NOW2000-Smirnov}
820: A.Yu. Smirnov, these proceedings.
821:
822: \bibitem{NOW2000-Kajita}
823: T. Kajita, these proceedings.
824:
825: \bibitem{NOW2000-Spentzouris}
826: P. Spentzouris, these proceedings.
827:
828: \bibitem{BGG-review-98-brief}
829: S.~M. Bilenky, C.~Giunti, and W.~Grimus,
830: \newblock Prog. Part. Nucl. Phys. {\bf 43}, 1 (1999).
831: %%CITATION = PPNPD,43,1;%%
832:
833: \bibitem{www-4nu}
834: Reference~list~at~www.to.infn.it/\~{}giunti/4nu.
835:
836: \bibitem{BGGS-AB-99-brief}
837: S.~M. Bilenky, C.~Giunti, W.~Grimus, and T.~Schwetz,
838: \newblock Phys. Rev. {\bf D60}, 073007 (1999).
839: %%CITATION = PHRVA,D60,073007;%%
840:
841: \bibitem{Barger-Fate-2000-brief}
842: V.~Barger~\emph{et~al.},~Phys.~Lett.~\textbf{B489},~345~(2000).
843:
844: \bibitem{Giunti-Laveder-ata-00}
845: C. Giunti and M. Laveder, hep-ph/0010009.
846:
847: \bibitem{Peres-Smirnov-3+1-00}
848: O. Peres and A. Smirnov, hep-ph/0011054.
849:
850: \bibitem{CHOOZ-99-brief}
851: M. Apollonio \textit{et al.}, Phys. Lett. \textbf{B466}, 415 (1999).
852:
853: \bibitem{Fogli-Lisi-Marrone-Scioscia-no3-99}
854: G. L. Fogli \textit{et al.}, hep-ph/9906450.
855:
856: \bibitem{Bugey-brief}
857: Y.~Declais~\emph{et~al.},~Nucl.~Phys.~\textbf{B434},~503~(1995).
858:
859: \bibitem{CDHS-brief}
860: F.~Dydak~\emph{et~al.},~Phys.~Lett.~\textbf{B134},~281~(1984).
861:
862: \bibitem{NOW2000-Steidl}
863: M. Steidl, these proceedings.
864:
865: \bibitem{BNL-E776}
866: L.~Borodovsky {\em et~al.},
867: \newblock Phys. Rev. Lett. {\bf 68}, 274 (1992).
868: %%CITATION = PRLTA,68,274;%%
869:
870: \bibitem{NOW2000-Zucchelli}
871: P. Zucchelli, these proceedings.
872:
873: \bibitem{NOW2000-Vidal-Sitjes}
874: G. Vidal-Sitjes, these proceedings.
875:
876: \bibitem{NOW2000-Suzuki}
877: Y. Suzuki, these proceedings.
878:
879: \bibitem{NOW2000-Ronga}
880: F. Ronga, these proceedings.
881:
882: \bibitem{Giunti-JHEP-00-brief}
883: C. Giunti, JHEP \textbf{0001}, 032 (2000).
884:
885: \bibitem{BGG-bounds-98-brief}
886: S.M. Bilenky, C. Giunti and W. Grimus, Phys. Rev. \textbf{D57}, 1920 (1998).
887:
888: \bibitem{Concha-SNO-00}
889: M.C. Gonzalez-Garcia and C. Pena-Garay, hep-ph/0011245.
890:
891: \bibitem{Yasuda-fouratm-00}
892: O. Yasuda, hep-ph/0006319.
893:
894: \bibitem{Fogli-Lisi-Marrone-fouratm-00-brief}
895: G.L. Fogli \textit{et al.}, hep-ph/0009299.
896:
897: \bibitem{Okada-Yasuda-97-brief}
898: N. Okada and O. Yasuda, Int. J. Mod. Phys. A \textbf{12}, 3669 (1997).
899:
900: \bibitem{Ibarra-Navarro-4nu-00-brief}
901: A. Ibarra and I. Navarro, JHEP \textbf{0002}, 031 (2000).
902:
903: \bibitem{Giunti-Neutrinoless-99-brief}
904: C. Giunti, Phys. Rev. \textbf{D61}, 036002 (2000).
905:
906: \end{thebibliography}
907:
908: \end{document}
909: