1: \documentstyle[prd,aps,preprint,tighten,epsfig]{revtex}
2:
3: \begin{document}
4:
5: \draft
6:
7: \title{Vanishing Effective Mass of the Neutrinoless Double Beta Decay?}
8: \author{{\bf Zhi-zhong Xing}}
9: \address{CCAST (World Laboratory), P.O. Box 8730, Beijing 100080, China \\
10: and Institute of High Energy Physics, Chinese Academy of Sciences, \\
11: P.O. Box 918 (4), Beijing 100039, China
12: \footnote{Mailing address} \\
13: ({\it Electronic address: xingzz@mail.ihep.ac.cn}) }
14: \maketitle
15:
16: \begin{abstract}
17: We stress the point that massive neutrinos may be Majorana particles
18: even if the effective mass of the neutrinoless double beta decay
19: $\langle m\rangle_{ee}$ vanishes. We show that current neutrino
20: oscillation data {\it do} allow $\langle m\rangle_{ee} = 0$ to hold,
21: if the Majorana CP-violating phases lie in two specific regions.
22: Strong constraints on the neutrino mass spectrum can then be obtained.
23: A possible texture of the neutrino mass matrix is also illustrated
24: under the $\langle m\rangle_{ee} = 0$ condition.
25: \end{abstract}
26:
27: \pacs{PACS number(s): 14.60.Pq, 13.10.+q, 25.30.Pt}
28:
29: \newpage
30:
31: The recent SK \cite{SK}, SNO \cite{SNO},
32: KamLAND \cite{KM} and K2K \cite{K2K} experiments have
33: provided us with very convincing evidence that the solar and
34: atmospheric neutrino anomalies are both due to neutrino
35: oscillations. The occurence of neutrino oscillations implies
36: that neutrinos are massive and lepton flavors are mixed. If
37: neutrinos are Majorana particles, a complete parametrization
38: of the $3\times 3$ lepton flavor mixing matrix requires
39: three mixing angles, one Dirac-type CP-violating phase and
40: two Majorana-type CP-violating phases \cite{FX01}. While
41: three mixing angles have been determined or constrained by
42: current neutrino oscillation data to an acceptable degree
43: of accuracy, three CP-violating phases are entirely
44: unrestricted. It is expected that the Dirac phase can be measured
45: from CP- or T-violating effects in the long-baseline neutrino
46: oscillation experiments \cite{LBL}. To measure two Majorana
47: phases is extremely difficult, because all possible
48: lepton-number-nonconserving processes induced by light Majorana
49: neutrinos are strongly suppressed in magnitude \cite{Kayser}.
50:
51: The most sensitive way to get some information on
52: two Majorana CP-violating phases is to detect the neutrinoless
53: double beta decay of some even-even nuclei,
54: \begin{equation}
55: A(Z,N) \; \rightarrow \; A(Z+2, N-2) + 2e^- \; ,
56: % (1)
57: \end{equation}
58: which can occur through the exchange of a Majorana neutrino between
59: two decaying neutrons inside a nucleus, as illustrated in Fig. 1.
60: It would be forbidden, however, if neutrinos were Dirac particles.
61: Thus the neutrinoless double beta decay provides us with a unique
62: opportunity to identify the Majorana nature of massive neutrinos.
63: The rate of the neutrinoless double beta decay is proportional to
64: an effective neutrino mass term, defined as
65: \begin{equation}
66: \langle m \rangle_{ee} \; = \; \left |
67: m_1 V^2_{e1} + m_2 V^2_{e2} + m_3 V^2_{e3} \right | \; ,
68: % (2)
69: \end{equation}
70: where $m_i$ (for $i=1,2,3$) denote the physical masses of three
71: neutrinos, and $V_{ei}$ stand for the elements in the first row
72: of the $3\times 3$ lepton flavor mixing matrix $V$. It is obvious
73: that $\langle m\rangle_{ee} = 0$ would trivially hold, if $m_i = 0$
74: were taken.
75:
76: While $\langle m\rangle_{ee} \neq 0$ must imply that neutrinos
77: are Majorana particles, $\langle m\rangle_{ee} = 0$ does not
78: {\it necessarily} imply that neutrinos are Dirac particles. The
79: reason is simply that the Majorana phases hidden in $V_{ei}$ may
80: lead to significant cancellations on the right-hand side of
81: Eq. (2), making $\langle m\rangle_{ee}$ vanishing or too small to
82: be detectable \cite{B}. Hence much care has to be taken, if no
83: convincing signal of the neutrinoless double beta decay can be
84: experimentally established
85: %%%%%%%%%%%%%%%%%%%%%%%%%%%
86: \footnote{Klapdor-Kleingrothaus {\it et al.} have recently reported
87: the first evidence for the neutrinoless double beta decay \cite{KK}.
88: However, their result was criticized by some authors \cite{KK2}.
89: Future experiments will have sufficiently high sensitivity to
90: clarify the present debates \cite{KK3}, to confirm or to
91: disprove the alleged result in Ref. \cite{KK}.}:
92: %%%%%%%%%%%%%%%%%%%%%%%%%%%
93: it may imply that (1) the experimental sensitivity is not high
94: enough; (2) the massive neutrinos are Dirac particles; or (3) the
95: vanishing or suppression of $\langle m\rangle_{ee}$ is due to
96: large cancellations induced by the Majorana CP-violating phases.
97: The third possibility is certainly interesting and
98: important \cite{Vissani}, and it deserves to be carefully examined
99: from a model-independent point of view and with the help of the
100: latest experimental data.
101:
102: The main purpose of this paper is to find out the parameter
103: space of two Majorana phases in the case of
104: $\langle m\rangle_{ee} = 0$. We demonstrate that current
105: neutrino oscillation data {\it do} allow
106: $\langle m\rangle_{ee} = 0$ to hold, if the Majorana phases lie
107: in two specific regions. Very strong constraints on three neutrino
108: masses can then be obtained. We find that the neutrino mass
109: spectrum performs a normal hierarchy: $m_1 < m_2 < m_3$. Finally,
110: we present some brief discussions about how to recast the texture
111: of the neutrino mass matrix under the $\langle m\rangle_{ee} = 0$
112: condition.
113:
114: It is clear in Eq. (2) that only the flavor mixing matrix elements
115: $V_{e1}$, $V_{e2}$ and $V_{e3}$ are relevant to the effective mass
116: of the neutrinoless double beta decay. Without loss of generality,
117: one may redefine the phases of three charged lepton fields in an
118: appropriate way such that the phases of $V_{e1}$ and $V_{e2}$ are
119: purely of the Majorana type and $V_{e3}$ is real \cite{Xing02}.
120: In other words,
121: \begin{equation}
122: \arg (V_{e1}) = \rho \; , ~~~~
123: \arg (V_{e2}) = \sigma \; , ~~~~
124: \arg (V_{e3}) = 0 \; .
125: % (3)
126: \end{equation}
127: Note that $\rho$ and $\sigma$ have nothing to do with CP and T
128: violation in normal neutrino oscillations. Taking account of
129: Eqs. (2) and (3), we find that
130: $\langle m\rangle_{ee} = 0$ requires
131: \begin{eqnarray}
132: && m_1 |V_{e1}|^2 \sin 2\rho + m_2 |V_{e2}|^2 \sin 2\sigma
133: \; =\; 0 \; , ~~~~~~
134: \nonumber \\
135: ~~~~~~~~ && m_1 |V_{e1}|^2 \cos 2\rho + m_2 |V_{e2}|^2 \cos 2\sigma
136: + m_3 |V_{e3}|^2 \; =\; 0 \; . ~~~~~~
137: % (4)
138: \end{eqnarray}
139: These two conditions, together with current experimental data
140: on the flavor mixing matrix elements ($|V_{e1}|$, $|V_{e2}|$
141: and $|V_{e3}|$) and the mass-squared differences of solar and
142: atmospheric neutrino oscillations
143: ($\Delta m^2_{\rm sun} \equiv |m^2_2 - m^2_1|$ and
144: $\Delta m^2_{\rm atm} \equiv |m^2_3 - m^2_2|$), allow us to
145: determine or constrain both the masses of three neutrinos
146: ($m_1$, $m_2$ and $m_3$) and the Majorana phases of CP violation
147: ($\rho$ and $\sigma$). More specific discussions about the
148: consequences of Eq. (4) are in order.
149:
150: (a) Note that the present solar neutrino data support
151: $0\leq m_1 < m_2$ \cite{SNO,KM}. If $m_1 =0$ holds, then Eq. (4)
152: requires $\sigma = (2n+1)\pi/2$ with $n=0,1,2$ $\cdot\cdot\cdot$ for
153: arbitrary values of $\rho$. In this case, we immediately obtain
154: \begin{equation}
155: \frac{m_2}{m_3} \; =\; \frac{|V_{e3}|^2}{|V_{e2}|^2} \;\; , ~~~~~~~
156: \frac{\Delta m^2_{\rm sun}}{\Delta m^2_{\rm atm}} \; =\;
157: \frac{|V_{e3}|^4}{|V_{e2}|^4 - |V_{e3}|^4} \;\; .
158: % (5)
159: \end{equation}
160: Because of $|V_{e3}| < |V_{e2}|$, the neutrino mass spectrum
161: performs a normal hierarchy (i.e., $m_3 > m_2 > m_1 =0$). The
162: absolute values of $m_2$ and $m_3$ are given by
163: \begin{equation}
164: ~~ m_2 = \frac{|V_{e3}|^2 \sqrt{\Delta m^2_{\rm atm}}}
165: {\sqrt{|V_{e2}|^4 - |V_{e3}|^4}} \;\; , ~~~
166: m_3 = \frac{|V_{e2}|^2 \sqrt{\Delta m^2_{\rm atm}}}
167: {\sqrt{|V_{e2}|^4 - |V_{e3}|^4}} \;\; . ~~
168: % (6)
169: \end{equation}
170: Our numerical analysis will show that the possibility
171: $m_1 = \langle m\rangle_{ee} =0$ is actually allowed by current
172: neutrino oscillation data.
173:
174: (b) Whether $0< m_2 < m_3$ or $0 \leq m_3 < m_2$ holds remains
175: an open question. If $m_3 = 0$ held, then Eq. (4) would require
176: $|\rho - \sigma| = (2n+1)\pi/2$ with $n$ being an arbitrary
177: integer. In this case, we would be led to
178: \begin{equation}
179: \frac{m_1}{m_2} \; =\; \frac{|V_{e2}|^2}{|V_{e1}|^2} \;\; , ~~~~~~~
180: \frac{\Delta m^2_{\rm sun}}{\Delta m^2_{\rm atm}} \; =\;
181: \frac{|V_{e1}|^4 - |V_{e2}|^4}{|V_{e1}|^4} \;\; ;
182: % (7)
183: \end{equation}
184: as well as
185: \begin{equation}
186: ~~ m_1 = \frac{|V_{e2}|^2 \sqrt{\Delta m^2_{\rm sun}}}
187: {\sqrt{|V_{e1}|^4 - |V_{e2}|^4}} \;\; , ~~~
188: m_2 = \frac{|V_{e1}|^2 \sqrt{\Delta m^2_{\rm sun}}}
189: {\sqrt{|V_{e1}|^4 - |V_{e2}|^4}} \;\; . ~~
190: % (8)
191: \end{equation}
192: Our numerical analysis will show that the possibility
193: $m_3 = \langle m\rangle_{ee} = 0$ has definitely been ruled
194: out by current neutrino oscillation data.
195:
196: (c) If $m_1 \neq 0$ but $\sin 2\rho =0$ or $\sin 2\sigma = 0$
197: holds, then Eq. (4) requires $\rho = n\pi$ and
198: $\sigma = (2n'+1)\pi/2$ with
199: \begin{equation}
200: m_1 |V_{e1}|^2 + m_3 |V_{e3}|^2 \; =\; m_2 |V_{e2}|^2 \; ;
201: % (9)
202: \end{equation}
203: or $\rho = (2n+1)\pi/2$ and $\sigma = n'\pi$ with
204: \begin{equation}
205: m_2 |V_{e2}|^2 + m_3 |V_{e3}|^2 \; =\; m_1 |V_{e1}|^2 \; ;
206: % (10)
207: \end{equation}
208: or $\rho = (2n+1)\pi/2$ and $\sigma = (2n'+1)\pi/2$ with
209: \begin{equation}
210: m_1 |V_{e1}|^2 + m_2 |V_{e2}|^2 \; =\; m_3 |V_{e3}|^2 \; ,
211: % (11)
212: \end{equation}
213: where $n$ and $n'$ are arbitrary integers. With the help of
214: \begin{eqnarray}
215: m_1 & = & \sqrt{m^2_3 \pm \Delta m^2_{\rm atm} +
216: \Delta m^2_{\rm sun}} \;\; , ~~~~
217: \nonumber \\
218: m_2 & = & \sqrt{m^2_3 \pm \Delta m^2_{\rm atm}} \;\; ,
219: % (12)
220: \end{eqnarray}
221: $m_3$ can be solved from Eq. (9), (10) or (11). Thus the
222: spectrum of three neutrino masses is fully determinable.
223: Our numerical analysis will show that Eqs. (9) and (10) are
224: consistent quite well with current neutrino oscillation data.
225: In comparison, Eq. (11) is also allowed but it is less favored.
226:
227: (d) Besides the special cases considered above, more general
228: results for $(m_1, m_2, m_3)$ and $(\rho, \sigma)$ can be
229: obtained from Eq. (4). Indeed,
230: \begin{eqnarray}
231: \frac{m_1}{m_2} & = & -\frac{|V_{e2}|^2}{|V_{e1}|^2}
232: \cdot \frac{\sin 2\sigma}{\sin 2\rho} \;\; ,
233: \nonumber \\
234: \frac{m_2}{m_3} & = & +\frac{|V_{e3}|^2}{|V_{e2}|^2}
235: \cdot \frac{\sin 2\rho}{\sin 2 (\sigma - \rho)} \;\; .
236: % (13)
237: \end{eqnarray}
238: Since $0 \leq m_1/m_2 <1$ and $0 < m_2/m_3$ hold, part of the
239: $(\rho, \sigma)$ parameter space must be excluded. Furthermore,
240: we arrive at
241: \begin{equation}
242: \frac{\Delta m^2_{\rm sun}}{\Delta m^2_{\rm atm}}
243: \; =\; \frac{|V_{e3}|^4}{|V_{e1}|^4} \cdot
244: \frac{\displaystyle \left | |V_{e1}|^4 \sin^2 2\rho -
245: |V_{e2}|^4 \sin^2 2\sigma \right |}
246: {\displaystyle \left | |V_{e2}|^4 \sin^2 2(\sigma -\rho) -
247: |V_{e3}|^4 \sin^2 2\rho \right |} \;\; .
248: % (14)
249: \end{equation}
250: Note that the lepton flavor mixing matrix elements $|V_{e1}|^2$,
251: $|V_{e2}|^2$, $|V_{e3}|^2$, $|V_{\mu 3}|^2$ and $|V_{\tau 3}|^2$
252: are associated respectively with the mixing factors of solar,
253: atmospheric and CHOOZ \cite{CHOOZ} reactor neutrino oscillations,
254: \begin{eqnarray}
255: \sin^2 2\theta_{\rm sun} & = & 4 |V_{e1}|^2 |V_{e2}|^2 \; ,
256: \nonumber \\
257: \sin^2 2\theta_{\rm atm} & = & 4 |V_{\mu 3}|^2
258: \left (1 - |V_{\mu 3}|^2 \right ) \; ,
259: \nonumber \\
260: \sin^2 2\theta_{\rm chz} & = & 4 |V_{e3}|^2
261: \left (1 - |V_{e3}|^2 \right ) \; .
262: % (15)
263: \end{eqnarray}
264: Reversely, we have \cite{Xing03}
265: \begin{eqnarray}
266: |V_{e1}|^2 & = & \frac{1}{2} \left ( \cos^2\theta_{\rm chz}
267: + \sqrt{\cos^4\theta_{\rm chz} - \sin^2 2\theta_{\rm sun}} \right ) \; ,
268: \nonumber \\
269: |V_{e2}|^2 & = & \frac{1}{2} \left ( \cos^2\theta_{\rm chz}
270: - \sqrt{\cos^4\theta_{\rm chz} - \sin^2 2\theta_{\rm sun}} \right ) \; ,
271: \nonumber \\
272: |V_{e3}|^2 & = & \sin^2 \theta_{\rm chz} \; ,
273: \nonumber \\
274: |V_{\mu 3}|^2 & = & \sin^2 \theta_{\rm atm} \; ,
275: \nonumber \\
276: |V_{\tau 3}|^2 & = & \cos^2 \theta_{\rm chz} -
277: \sin^2 \theta_{\rm atm} \; .
278: % (16)
279: \end{eqnarray}
280: In view of the recent SK \cite{SK}, SNO \cite{SNO},
281: KamLAND \cite{KM}, K2K \cite{K2K} and CHOOZ \cite{CHOOZ}
282: data on neutrino oscillations, we have
283: $\Delta m^2_{\rm sun} \in [5.9, ~ 8.8] \times 10^{-5} ~ {\rm eV}^2$,
284: $\sin^2 \theta_{\rm sun} \in [0.25, ~ 0.40]$ \cite{Fit};
285: $\Delta m^2_{\rm atm} \in [1.65, ~ 3.25] \times 10^{-3} ~ {\rm eV}^2$,
286: $\sin^2 2\theta_{\rm atm} \in [0.88, ~ 1.00]$ \cite{Fogli}; and
287: $\sin^2 2\theta_{\rm chz} < 0.2$ at the $90\%$ confidence level.
288: With the help of these experimental results, the allowed ranges
289: of $\rho$ and $\sigma$ can then be obtained from Eqs. (14) and (16).
290:
291: We plot the parameter space of two Majorana phases in Fig. 2(a).
292: It becomes obvious that $\rho$ and $\sigma$ may take many nontrivial
293: values, which guarantee $\langle m\rangle_{ee} =0$. This important
294: point has not been observed before. Note that the $(\rho, \sigma)$
295: parameter space in Fig. 2(a) can be generalized to the larger
296: $(\rho \pm n_1\pi, \sigma \pm n_2\pi)$ parameter space, where $n_1$
297: and $n_2$ are arbitrary integers. For illustration, we typically pick
298: \begin{equation}
299: (\rho, ~\sigma) \; = \; \left (\frac{\pi}{4}, ~\frac{2\pi}{3} \right)
300: \;\;\;\;\; {\rm or} \;\;\;\;\; \left (\frac{3\pi}{4}, ~\frac{\pi}{3} \right)
301: % (17)
302: \end{equation}
303: from Fig. 2(a). Then we arrive at
304: \begin{equation}
305: \frac{m_1}{m_2} \; = \;
306: \frac{\sqrt{3}}{2} \frac{|V_{e2}|^2}{|V_{e1}|^2} \;\; ,
307: ~~~~~
308: \frac{m_2}{m_3} \; = \; 2 \frac{|V_{e3}|^2}{|V_{e2}|^2} \;\; .
309: % (18)
310: \end{equation}
311: Given the ranges of $\theta_{\rm sun}$ and $\theta_{\rm chz}$ favored
312: by current experimental data, $m_1 < m_2$ and $m_2 < m_3$ are found to
313: hold from Eq. (18). Thus three neutrino masses perform a normal
314: hierarchy in this specific but interesting case.
315:
316: A detailed analysis of the $(m_1/m_2, m_2/m_3)$ parameter space is
317: shown in Fig. 2(b). We find that two specific regions of
318: $\rho$ and $\sigma$ in Fig. 2(a) correspond to a common region
319: of $m_1/m_2$ and $m_2/m_3$ in Fig. 2(b), just like the specific
320: case illustrated in Eqs. (17) and (18). One can see that both
321: $0\leq m_1/m_2 <1$ and $0< m_2/m_3 <1$ hold,
322: thus three neutrino masses have a normal hierarchy
323: (i.e., $m_1 < m_2 < m_3$). When $m_1/m_2$ approaches 1,
324: $m_2/m_3$ is somehow close to 1 too. In this case, which is not
325: very likely, three neutrino masses are nearly degenerate
326: (i.e., $m_1 \approx m_2 \approx m_3$). Note that $m_2/m_3$ and
327: $|V_{e3}|$ have minimal values $(m_2/m_3)_{\rm min} \approx 0.135$
328: and $(|V_{e3}|)_{\rm min} \approx 0.0695$, respectively. The
329: latter, which corresponds to
330: $(\theta_{\rm chz})_{\rm min} \approx 4^\circ$
331: or $(\sin^2 2\theta_{\rm chz})_{\rm min} \approx 0.02$,
332: is associated with the prerequisite $\langle m\rangle_{ee} =0$.
333: In contrast, we do not find any restriction on the input and
334: output values of $\theta_{\rm sun}$ or $\theta_{\rm atm}$ in
335: our numerical calculation.
336:
337: We see that $m_2/m_3$ is most likely to be around 0.2, implying
338: that $m_3 \approx \sqrt{\Delta m^2_{\rm atm}} \sim 0.05 ~ {\rm eV}$
339: should be a good approximation. In this case, the effective mass
340: of the tritium beta decay $\langle m\rangle_e$ is too small to
341: be detected by the KATRIN experiment \cite{No}. If
342: $m_2/m_3 \sim 0.9$ is taken, one will get $m_3 \sim 0.1$ eV and
343: $m_1 + m_2 + m_3 \sim 0.3$ eV, consistent with the recent WMAP data
344: $m_1 + m_2 + m_3 < 0.71$ eV \cite{WMAP}.
345:
346: Let us proceed to discuss how to recast the neutrino mass matrix
347: under the condition $\langle m\rangle_{ee} =0$. In the flavor basis
348: where the charged lepton mass matrix is diagonal, the neutrino
349: mass matrix $M$ can be written as
350: \begin{equation}
351: M \; =\; V \left ( \matrix{
352: m_1 & 0 & 0 \cr
353: 0 & m_2 & 0 \cr
354: 0 & 0 & m_3 \cr} \right ) V^T \; .
355: % (19)
356: \end{equation}
357: The lepton flavor mixing matrix $V$ consists of three nontrivial
358: CP-violating phases: the Dirac phase $\delta$ and the Majorana
359: phases $\rho$ and $\sigma$. For simplicity, we assume $\delta =0$
360: to examine the dependence of $M$ on $\rho$ and $\sigma$. Then
361: $V$ may take the form
362: \begin{equation}
363: V \; =\; \left ( \matrix{
364: |V_{e1}| & |V_{e2}| & |V_{e3}| \cr
365: -|V_{\mu 1}| & |V_{\mu 2}| & |V_{\mu 3}| \cr
366: |V_{\tau 1}| & -|V_{\tau 2}| & |V_{\tau 3}| \cr}
367: \right )
368: \left ( \matrix{
369: e^{i\rho} & 0 & 0 \cr
370: 0 & e^{i\sigma} & 0 \cr
371: 0 & 0 & 1 \cr} \right ) \; ,
372: % (20)
373: \end{equation}
374: in which $|V_{e1}|$, $|V_{e2}|$, $|V_{e3}|$, $|V_{\mu 3}|$ and
375: $|V_{\tau 3}|$ have been given in Eq. (16) in terms of
376: $\theta_{\rm sun}$, $\theta_{\rm atm}$ and $\theta_{\rm chz}$;
377: and $|V_{\mu 1}|$, $|V_{\mu 2}|$, $|V_{\tau 1}|$ and $|V_{\tau 2}|$
378: read as follows (in the assumption of $\delta =0$ \cite{Xing03}):
379: \begin{eqnarray}
380: |V_{\mu 1}| & = & \frac{|V_{e2}| |V_{\tau 3}| + |V_{e1}|
381: |V_{e3}| |V_{\mu 3}|}{1 - |V_{e3}|^2} \;\; ,
382: \nonumber \\
383: |V_{\mu 2}| & = & \frac{|V_{e1}| |V_{\tau 3}| - |V_{e2}|
384: |V_{e3}| |V_{\mu 3}|}{1 - |V_{e3}|^2} \;\; ,
385: \nonumber \\
386: |V_{\tau 1}| & = & \frac{|V_{e2}| |V_{\mu 3}| - |V_{e1}|
387: |V_{e3}| |V_{\tau 3}|}{1 - |V_{e3}|^2} \;\; ,
388: \nonumber \\
389: |V_{\tau 2}| & = & \frac{|V_{e1}| |V_{\mu 3}| + |V_{e2}|
390: |V_{e3}| |V_{\tau 3}|}{1 - |V_{e3}|^2} \;\; .
391: % (21)
392: \end{eqnarray}
393: With the help of Eqs. (19) and (21), six independent elements of
394: the symmetric neutrino mass matrix $M$ can be expressed as
395: \begin{equation}
396: \left ( \matrix{
397: M_{ee} \cr M_{e \mu} \cr M_{e \tau} \cr M_{\mu \mu} \cr
398: M_{\mu \tau} \cr M_{\tau \tau} \cr} \right ) \; =\;
399: m_1 \left ( \matrix{
400: |V_{e1}|^2 \cr -|V_{e1}| |V_{\mu 1}| \cr |V_{e1}| |V_{\tau 1}| \cr
401: |V_{\mu 1}|^2 \cr -|V_{\mu 1}| |V_{\tau 1}| \cr |V_{\tau 1}|^2 \cr}
402: \right ) e^{2i\rho} + m_2 \left ( \matrix{
403: |V_{e2}|^2 \cr |V_{e2}| |V_{\mu 2}| \cr -|V_{e2}| |V_{\tau 2}| \cr
404: |V_{\mu 2}|^2 \cr -|V_{\mu 2}| |V_{\tau 2}| \cr |V_{\tau 2}|^2 \cr}
405: \right ) e^{2i\sigma} + m_3 \left ( \matrix{
406: |V_{e3}|^2 \cr |V_{e3}| |V_{\mu 3}| \cr |V_{e3}| |V_{\tau 3}| \cr
407: |V_{\mu 3}|^2 \cr |V_{\mu 3}| |V_{\tau 3}| \cr |V_{\tau 3}|^2 \cr}
408: \right ) \; .
409: % (22)
410: \end{equation}
411: As $|M_{ee}| = \langle m\rangle_{ee}$ holds, we have $M_{ee} =0$
412: under the condition $\langle m\rangle_{ee} =0$. For illustration,
413: we calculate the other five matrix elements in Eq. (22) by using
414: the typical values of $\rho$ and $\sigma$ chosen in Eq. (17) and
415: taking $\theta_{\rm sun} = 33^\circ$, $\theta_{\rm atm} = 45^\circ$
416: and $\theta_{\rm chz} = 9^\circ$. The numerical result are
417: $m_1/m_2 \approx 0.42$, $m_2/m_3 \approx 0.15$, and
418: \begin{equation}
419: M \; \approx \; m_3 \left ( \matrix{
420: {\bf 0} & 0.11 e^{\mp i 36^\circ} & 0.15 e^{\pm i 26^\circ} \cr
421: 0.11 e^{\mp i 36^\circ} & 0.48 e^{\mp i 2.3^\circ} &
422: 0.51 e^{\pm i 3.8^\circ} \cr
423: 0.15 e^{\pm i 26^\circ} & 0.51 e^{\pm i 3.8^\circ} &
424: 0.45 e^{\mp i 6.5^\circ} \cr} \right ) \; ,
425: % (23)
426: \end{equation}
427: where $m_3 \sim 0.05$ eV. We see that this one-zero texture of
428: $M$ does not perform an apparent hierarchy and has little
429: similarity with those two-zero textures of $M$ illustrated in
430: Ref. \cite{Xing02a}. Of course, the form of $M$ in Eq. (23) will
431: get modified, if the Dirac CP-violating phase of $V$ is switched
432: on. A more general and delicate analysis of possible patterns of
433: $M$ under the $\langle m\rangle_{ee} =0$ condition will be done
434: elsewhere.
435:
436: It is worth mentioning that the $\langle m\rangle_{ee} =0$
437: condition may be taken as a prerequisite to build models
438: for the Majorana neutrino mass matrix $M$, if the neutrinoless
439: double beta decay is unable to be detected. Another two empirical
440: conditions, ${\rm Det}M =0$ \cite{Yanagida} and
441: $\lambda_1 + \lambda_2 + \lambda_3 =0$ (where
442: $|\lambda_i| = m_i$ \cite{He}), have also been discussed to
443: constrain the form of $M$ in a phenomenological way. Such
444: ``zero'' conditions are similar to those taken for the matrix
445: elements of $M$ \cite{Xing02a,F}, in order to reduce the number of
446: free parameters in $M$. This is one of a few realistic
447: approaches \cite{Review}, towards some deeper understanding of
448: lepton mass generation and flavor mixing.
449:
450: In summary, we have stressed the point that
451: $\langle m\rangle_{ee} =0$ does not necessarily indicate the
452: Dirac nature of light neutrinos. Current neutrino oscillation
453: data do allow $\langle m\rangle_{ee} =0$ to hold, if neutrinos
454: are Majorana particles and their two Majorana CP-violating
455: phases lie in two specific regions. This observation will be
456: important for model building, in order to completely understand
457: or interpret current and future experimental data on the neutrino
458: mass spectrum and the neutrinoless double beta decay.
459:
460: \vspace{0.5cm}
461:
462: I am grateful to C. Giunti and F. Vissani for useful comments
463: and discussions, and to F. Vissani and W. Rodejohann for bringing
464: a few relevant references to my attention. I am also indebted to
465: W.L. Guo for his help in dealing with the figures. This work was
466: supported in part by the National Natural Science Foundation of China.
467:
468: \newpage
469:
470: \begin{thebibliography}{99}
471: \bibitem{SK} For a review, see: C.K. Jung, C. McGrew, T. Kajita,
472: and T. Mann, Ann. Rev. Nucl. Part. Sci. {\bf 51}, 451 (2001).
473:
474: \bibitem{SNO} SNO Collaboration, Q.R. Ahmad {\it et al.},
475: Phys. Rev. Lett. {\bf 89}, 011301 (2002); Phys. Rev. Lett.
476: {\bf 89}, 011302 (2002).
477:
478: \bibitem{KM} KamLAND Collaboration, K. Eguchi {\it et al.},
479: Phys. Rev. Lett. {\bf 90}, 021802 (2003).
480:
481: \bibitem{K2K} K2K Collaboration, M.H. Ahn {\it et al.},
482: Phys. Rev. Lett. {\bf 90}, 041801 (2003).
483:
484: \bibitem{FX01} H. Fritzsch and Z.Z. Xing,
485: Phys. Lett. B {\bf 517}, 363 (2001); and references therein.
486:
487: \bibitem{LBL} See, e.g., A. Blondel {\it et al.},
488: Nucl. Instrum. Meth. A {\bf 451}, 102 (2000);
489: C. Albright {\it et al.}, hep-ex/0008064;
490: H. Fritzsch and Z.Z. Xing, Phys. Rev. D {\bf 61}, 073016 (2000);
491: G. Barenboim {\it et al.}, hep-ex/0304017.
492:
493: \bibitem{Kayser} V. Barger, S.L. Glashow, P. Langacker, and
494: D. Marfatia, Phys. Lett. B {\bf 540}, 247 (2002);
495: A. de Gouvea, B. Kayser, and R.N. Mohapatra,
496: Phys. Rev. D {\bf 67}, 053004 (2003).
497:
498: \bibitem{B} S.M. Bilenky, S. Pascoli, and S.T. Petcov,
499: Phys. Rev. D {\bf 64}, 053010 (2001);
500: W. Rodejohann, Nucl. Phys. B {\bf 597}, 110 (2001);
501: hep-ph/0203214; and references therein.
502:
503: \bibitem{KK} H.V. Klapdor-Kleingrothaus, A. Dietz, H.L. Harney,
504: and I.V. Krivosheina, Mod. Phys. Lett. A {\bf 16}, 2409 (2002).
505:
506: \bibitem{KK2} F. Feruglio, A. Strumia, and F. Vissani,
507: Nucl. Phys. B {\bf 637}, 345 (2002);
508: C.E. Aalseth {\it et al.}, Mod. Phys. Lett. A {\bf 17}, 1475 (2002);
509: Y.G. Zdesenko, F.A. Danevich, and V.I. Tretyak,
510: Phys. Lett. B {\bf 546}, 206 (2002).
511:
512: \bibitem{KK3} H.V. Klapdor-Kleingrothaus,
513: hep-ph/0302237; hep-ph/0303217; and references therein.
514:
515: \bibitem{Vissani} L. Wolfenstein,
516: in {\it Proc. of Neutrino 84} (1984), p. 730;
517: F. Vissani, JHEP {\bf 9906}, 022 (1999);
518: F. Feruglio, A. Strumia, and F. Vissani, in Ref. \cite{KK2}.
519:
520: \bibitem{Xing02} Z.Z. Xing, Phys. Rev. D {\bf 65}, 077302 (2002);
521: Phys. Rev. D {\bf 65}, 113010 (2002).
522:
523: \bibitem{CHOOZ} CHOOZ Collaboration, M. Apollonio {\it et al.},
524: Phys. Lett. B {\bf 420}, 397 (1998);
525: Palo Verde Collaboration, F. Boehm {\it et al.},
526: Phys. Rev. Lett. {\bf 84}, 3764 (2000).
527:
528: \bibitem{Xing03} W.L. Guo and Z.Z. Xing,
529: Phys. Rev. D {\bf 67}, 053002 (2003).
530:
531: \bibitem{Fit} V. Barger and D. Marfatia, hep-ph/0212126;
532: G.L. Fogli {\it et al.}, hep-ph/0212127;
533: M. Maltoni, T. Schwetz, and J.W.F. Valle, hep-ph/0212129;
534: J.N. Bahcall, M.C. Gonzalez-Garcia, and C. Pena-Garay,
535: hep-ph/0212147;
536: P. Aliani, V. Antonelli, M. Picariello, and E. Torrente-Lujan,
537: hep-ph/0212212;
538: P.C. de Holanda and A.Yu. Smirnov, hep-ph/0212270.
539:
540: \bibitem{Fogli} G.L. Fogli, E. Lisi, A. Marrone, and D. Montanino,
541: hep-ph/0303064.
542:
543: \bibitem{No} Z.Z. Xing, hep-ph/0303178.
544:
545: \bibitem{WMAP} C.L. Bennett {\it et al.}, astro-ph/0302207;
546: D.N. Spergel {\it et al.}, astro-ph/0302209.
547:
548: \bibitem{Xing02a} Z.Z. Xing, Phys. Lett. B {\bf 530}, 159 (2002);
549: Phys. Lett. B {\bf 539}, 85 (2002).
550:
551: \bibitem{Yanagida} G.C. Branco, R. Felipe, F. Joaquim, and
552: T. Yanagida, hep-ph/0212341.
553:
554: \bibitem{He} X.G. He and A. Zee, hep-ph/0302201.
555:
556: \bibitem{F} P.H. Frampton, S.L. Glashow, and D. Marfatia,
557: Phys. Lett. B {\bf 536}, 79 (2002);
558: Z.Z. Xing, Phys. Lett. B {\bf 550}, 178 (2002).
559:
560: \bibitem{Review} For recent reviews with extensive references, see:
561: H. Fritzsch and Z.Z. Xing, Prog. Part. Nucl. Phys. {\bf 45}, 1 (2000);
562: G. Altarelli and F. Feruglio, hep-ph/0206077, to appear in
563: {\it Neutrino Mass} - Springer Tracts in Modern Physics, edited by
564: G. Altarelli and K. Winter (2002).
565:
566: \end{thebibliography}
567:
568: \newpage
569:
570: %%%%%%%%%%%%%%%%%%%% Fig. 1 %%%%%%%%%%%%%%%%
571: \begin{figure}[t]
572: \vspace{-1.2cm}
573: \epsfig{file=bbf1.ps,bbllx=3.5cm,bblly=15cm,bburx=18cm,bbury=32.5cm,%
574: width=15.5cm,height=22cm,angle=0,clip=}
575: \vspace{-9.8cm}
576: \caption{Illustrative plot for the neutrinoless double beta decay
577: of some even-even nuclei via the exchange of a virtual Majorana
578: neutrino.}
579: \end{figure}
580: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
581:
582: \newpage
583:
584: %%%%%%%%%%%%%%%%%%%% Fig. 2 %%%%%%%%%%%%%%%%
585: \begin{figure}[t]
586: \vspace{-2cm}
587: \epsfig{file=bbf2m.ps,bbllx=-5cm,bblly=-2cm,bburx=18cm,bbury=26cm,%
588: width=15.5cm,height=20cm,angle=0,clip=}
589: \vspace{-0.8cm}
590: \caption{Implications of $\langle m\rangle_{ee} =0$:
591: the $(\rho, \sigma)$ and $(m_1/m_2, m_2/m_3)$ regions allowed
592: by current neutrino oscillation data.}
593: \end{figure}
594: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
595:
596:
597: \end{document}
598:
599:
600: