1: %New 10/25/05
2: \documentclass[12pt]{article}
3: \usepackage{graphicx}
4: \usepackage{epsfig}
5:
6: \newcommand{\spone}{0.9}
7: \newcommand{\sphalf}{1.15}
8: \newcommand{\sptwo}{1.4}
9: \newcommand{\spthree}{2.4}
10:
11: \newcommand{\singlespace}{\edef\baselinestretch{\spone}\Large\normalsize}
12: \newcommand{\kimspace}{\edef\baselinestretch{\sphalf}\Large\normalsize}
13: \newcommand{\doublespace}{\edef\baselinestretch{\sptwo}\Large\normalsize}
14: \newcommand{\threespace}{\edef\baselinestretch{\spthree}\Large\normalsize}
15:
16: %Choose one:
17:
18: \singlespace
19: \threespace
20: \doublespace
21:
22: \thispagestyle{empty}
23: \everymath={\displaystyle}
24: \begin{document}
25: \doublespace
26:
27:
28: \begin{center}
29: {\bf Rephasing Invariance and Hierarchy
30: of the CKM Matrix\\
31: $~$\\
32: T.K. Kuo$^*$ and Lu-Xin Liu$^\dagger$\\
33: Physics Department, Purdue University, W. Lafayette, IN 47907}
34: \end{center}
35: \vspace{30pt}
36:
37:
38: \begin{center}
39: {\it Abstract}
40: \end{center}
41:
42: We identified a set of four rephasing invariant parameters of the CKM
43: matrix. They are found to exhibit hierarchies in powers of $\lambda^2$,
44: from $\lambda^2$ to $\lambda^8$. It is shown that, at the present level
45: of accuracy, only the first three parameters are needed to fit all
46: available data on flavor physics.
47: \vspace{60pt}
48:
49: \noindent
50: $^*$E-mail: tkkuo@physics.purdue.edu
51:
52: \noindent
53: $^\dagger$E-mail: liul@physics.purdue.edu
54:
55: \pagebreak
56:
57: In the standard model, the esential input in understanding flavor
58: physics and CP violation is given by the Cabibbo-Kobayashi-Maskawa
59: (CKM) matrix, $V_{CKM}$, or rather, by the rephasing invariant
60: portions thereof, since one can always change the phases (rephasing)
61: of quark fields without affecting the physics. What is remarkable
62: is that, to this date, all observations, both CP-conserving and
63: CP-violating, while greatly over-constraining the parameters in
64: $V_{CKM}$, are nevertheless completely in accord with this picture.
65: Another remarkable feature of $V_{CKM}$ is its hierarchical structure,
66: characterized by the Wolfenstein parameter [1],
67: $\lambda = |V_{us}|\cong 0.22$. This implies that flavor processes can
68: be naturally classified by their strengths in powers of $\lambda$
69: (e.g. CP-violating phenomena only appear at
70: $O(\lambda^6)$ or higher). In this paper we make use of a
71: manifestly rephasing invariant parametrization introduced in a
72: previous work [2]. It consists of six parameters,
73: $(x_i,y_j), i,j =1,2,3$, which satisfy two constraints, so that
74: any four of them can be used as a complete set. Using known measurements,
75: we find that three of those $(y_2,y_1,x_2)$ are of order
76: $O(\lambda^2), O(\lambda^4)$ and $O(\lambda^6)$, respectively.
77: There is also considerable evidence that a fourth, $y_3$,
78: is of order $O(\lambda^8)$, and is consistent with zero, at the
79: present level of accuracy in the determination of $V_{CKM}$. It is then
80: shown that, for $V_{CKM}$, the set of three parameters
81: $(y_2,y_1,x_2)$, with $y_3=0$, suffices to describe all extant
82: data on flavor physics. An accurate assessment of the value of
83: $y_3$ can only be done after more precise measurements become
84: available.
85:
86: We begin by summarizing the main properties of the $(x,y)$ parametrization
87: and detailing its relations with other parametrizations which are
88: in common use.
89:
90: We assume, without loss of generosity, that
91: $$ det~V_{CKM} = + 1.
92: \eqno{(1)}
93: $$
94:
95: \noindent
96: There are then six rephasing invariants defined by
97: $$ \Gamma_{ijk} = V_{1i} V_{2j} V_{3k},
98: \eqno{(2)}
99: $$
100:
101: \noindent
102: where $(i,j,k)$ is a permutation of (1,2,3). It was proved
103: that all six $\Gamma_{ijk}$'s have the same imaginary part, $-iJ$,
104: where $J$ is the invariant CP measure [3], so that
105: $$ \Gamma_{ijk} = Re\Gamma_{ijk} - iJ.
106: \eqno{(3)}
107: $$
108: It is useful to separate the even and odd permutation $\Gamma$'s and
109: define
110: $$ Re(\Gamma_{123}, \Gamma_{231}, \Gamma_{312}) = (x_1,x_2,x_3);
111: \eqno{(4)}
112: $$
113: $$ Re(\Gamma_{132}, \Gamma_{213}, \Gamma_{321}) = (y_1,y_2,y_3).
114: \eqno{(5)}
115: $$
116: They are found to satisfy two constraints,
117: $$ (x_1 + x_2 + x_3) - (y_1 + y_2 + y_3) = 1,
118: \eqno{(6)}
119: $$
120: $$ x_1x_2 + x_2x_3 + x_3x_1 = y_1y_2 + y_2y_3 + y_3y_1 .
121: \eqno{(7)}
122: $$
123: Thus, any four of the set $(x_i,y_j)$ may be used as a complete set of
124: parameters of $V_{CKM}$. In addition, it was also established that
125: $$ J^2 = x_1x_2 x_3 - y_1y_2y_3.
126: \eqno{(8)}
127: $$
128: All of the parameters $(x_i,y_j)$ take values between
129: $\pm 1, -1 \leq (x_i,y_j) \leq +1$, with $y_j \leq x_i$, for
130: any $(i,j)$.
131:
132: We now turn to the relations between the set $(x,y)$ and other familiar
133: parametrizations. The simplest is that with $|V_{ij}|^2$ [4],
134: the absolute square of the elements of $V_{CKM}$. It is given by
135: $$ W = \left( \begin{array}{ccc}
136: |V_{11}|^2 & |V_{12}|^2 & |V_{13}|^2\\
137: |V_{21}|^2 & |V_{22}|^2 & |V_{23}|^2\\
138: |V_{31}|^2 & |V_{32}|^2 & |V_{33}|^2\\
139: \end{array} \right)
140: \eqno{(9)}
141: $$
142:
143: $$ = \left( \begin{array}{ccc}
144: x_1-y_1 & x_2-y_2 & x_3-y_3\\
145: x_3-y_2 & x_1-y_3 & x_2-y_1\\
146: x_2-y_3 & x_3-y_1 & x_1-y_2
147: \end{array} \right)
148: \eqno{(10)}
149: $$
150: It is also straight-forward to obtain the relations between
151: $(x,y)$ and the ``standard" parametrization of the Particle Data Group [5]
152: $$ V^{(s)} = \left( \begin{array}{ccc}
153: c_{12}c_{13} & s_{12} c_{13} & s_{13} e^{-i\delta}\\
154: -s_{12}c_{23}-c_{12}s_{23}s_{13}e^{i\delta} &
155: c_{12}c_{23}-s_{12}s_{23}s_{13}e^{i\delta} &
156: s_{23} c_{13} \\
157: s_{12}s_{23} -c_{12}c_{23}s_{13}e^{i\delta} &
158: -c_{12}s_{23}-s_{12}c_{23}s_{13}e^{i\delta} &
159: c_{23}c_{13} \end{array} \right) .
160: \eqno{(11)}
161: $$
162:
163: \noindent
164: Note that $det~V^{(s)} = +1$. However, $V^{(s)}$ is not invariant
165: under rephasing. We will not write down the explicit relations between
166: $(x,y)$ and the angles in $V^{(s)}$. These relations are exact but they
167: tend to be rather cumbersome.
168:
169: Another well-established parametrization,
170: due to Wolfenstein [1], is often used. It is an expansion of
171: $V_{CKM}$ in the parameter $\lambda = |V_{us}| \cong 0.22$,
172: whose validity is related to the fact that the three angles in
173: $V^{(s)}$ are small and hierarchical.
174: Wolfenstein's representation of $V_{CKM}$ reads:
175: $$ V^{(W)} = \left( \begin{array}{ccc}
176: 1-\lambda^2/2 & \lambda & A\lambda^3(\rho-i\eta) \\
177: -\lambda & 1-\lambda^2/2 & A\lambda^2 \\
178: A\lambda^3(1-\rho-i\eta) & -A\lambda^2 & 1 \end{array}\right)
179: + O(\lambda^4).
180: \eqno{(12)}
181: $$
182: Note that $det~V^{(W)} = 1 + O(\lambda^2)$, and, like $V^{(s)}$,
183: $V^{(W)}$ is not rephasing invariant. A systematic expansion
184: into higher orders of $\lambda$ is also available [6],
185: and is often used in the literature. In the spirit of an expansion
186: in $\lambda$, we may compute $(x,y)$ by using $\Gamma_{ijk}^{(W)} =
187: V_{1i}^{(W)} V_{2j}^{(W)} V_{3k}^{(W)}$, which form rephasing
188: invariant combinations. We find, to leading order in $\lambda^2$,
189: $$ \Gamma_{123}^{(W)} = 1 - \lambda^2 = x_1,
190: \eqno{(13)}
191: $$
192: $$ \Gamma_{213}^{(W)} = -\lambda^2 = y_2,
193: \eqno{(14)}
194: $$
195: $$ \Gamma_{132}^{(W)} = -A^2\lambda^4 = y_1.
196: \eqno{(15)}
197: $$
198: None of these $\Gamma$'s contains the imaginary part,
199: $-iJ$, which is $O(\lambda^6)$,
200: although they do give correct values for
201: $x_1,y_2$ and $y_1$, to leading order. On the other
202: hand,
203: $$ \Gamma_{231}^{(W)} = A^2\lambda^6 [(1-\rho)-i\eta] = x_2 -iJ,
204: \eqno{(16)}
205: $$
206: $$ \Gamma_{321}^{(W)} = A^2\lambda^6 [(\rho-\rho^2-\eta^2)-i\eta]
207: = y_3-iJ,
208: \eqno{(17)}
209: $$
210: $$\Gamma_{312}^{(W)} = A^2\lambda^6(\rho-i\eta) = x_3 -iJ.
211: \eqno{(18)}
212: $$
213:
214: \noindent
215: All of these quantities are $O(\lambda^6)$, and they have the
216: same imaginary part, with
217: $$ J = A^2\lambda^6 \eta,
218: \eqno{(19)}
219: $$
220: which is a well-known approximate result. Had we used the improved
221: version of $V^{(W)}$ [6], it will be seen that, except for
222: $\Gamma_{123}^{(W)}$, all $\Gamma^{(W)}$'s contain the same
223: imaginary part, $-iJ$.
224:
225: The above analysis shows that, to leading order in $\lambda^2$,
226: the $(x,y)$ parameters are given in the hierarchical order
227: $x_1 = O(1)$, $x_2 = O(\lambda^6)$, $x_3 = O(\lambda^6)$, $y_1 =
228: O(\lambda^4)$, $y_2 = O(\lambda^2)$ and $y_3 = O(\lambda^6)$. These
229: results are obtained with the tacit assumption that, in the
230: parametrization $V^{(W)}$,
231: the values $(A,\rho,\eta)$ are all of order one and are unrelated
232: to each other. When we incorporate recent measurements, it turns
233: out that there are correlations and detailed structures which have
234: interesting implications. This we will discuss in the following.
235:
236: We begin by improving upon our earlier analysis in Ref. [2], and
237: write down more precise relations between the $(x,y)$ parameters
238: and the angles $(\alpha,\beta,\gamma)$ of the unitarity triangle.
239: From the unitarity condition,
240: $$ V_{11}V_{13}^* + V_{21}V_{23}^* + V_{31}V_{33}^* = 0,
241: \eqno{(20)}
242: $$
243: we obtain a rephasing invariant equation by multiplying by
244: $V_{11}^*V_{13}$. The result is a triangle in the complex plane
245: whose height is (exactly) $iJ$, with one side given by
246: $$V_{11}^*V_{13}V_{31}V_{32}^* = y_3 + |V_{13}|^2|V_{31}|^2 - iJ.
247: \eqno{(21)}
248: $$
249: This triangle is plotted in Fig.1.
250:
251: Since $|V_{13}|^2 |V_{31}|^2 \leq O(\lambda^{12})$, together with
252: the estimate (to be discussed later) $ y\approx O(\lambda^8)$,
253: we have, from Fig. 1,
254:
255: \begin{figure}
256: \centering
257: \includegraphics[width=6cm,height=6cm]{secondgraph3.eps}
258: \caption{Rescaled unitarity triangle with base
259: $|V_{11}|^2 |V_{13}|^2$ and sides as labelled.
260: Also, $\Delta=|V_{13}|^2 |V_{31}|^2=O(\lambda^{12}),
261: \Delta '=|V_{13}|^2 |V_{21}|^2=O(\lambda^8).$}\label{fig:secondgraph3.eps}
262: \end{figure}
263:
264: $$ \tan \alpha =-(J/y_3) (1+ O(\lambda^4)).
265: \eqno{(22)}
266: $$
267: Similarly, using Ref.[2], we find
268: $$ \tan\beta = (J/x_2) (1+ O(\lambda^4)),
269: \eqno{(23)}
270: $$
271: $$\tan\gamma = (J/x_3) (1 + O(\lambda^2)).
272: \eqno{(24)}
273: $$
274:
275: \noindent
276: Thus, to a high degree of accuracy, the angles $(\alpha,\beta,\gamma)$
277: are the phase angles of $(\Gamma_{321}, \Gamma_{231}^*, \Gamma_{312}^*)$.
278: (Note that, in Ref.[2], Eq.(60) has a wrong sign, and the statement
279: thereafter contains a typo.)
280:
281: Experimentally, $\beta$ has been rather precisely measured,
282: while $\alpha$ and $\gamma$ are not so well determined.
283: To assess their effects on the $(x,y)$ parameters, we make
284: use of currently available global fits [7,8]. Because the
285: two analyses yield similar results, we will only quote the
286: values from Ref.[7], for simplicity.
287: The values of $(\alpha,\beta,\gamma)$, in degrees, are given by
288: $$\alpha = 94_{-10}^{+12},~~ \beta = 24 \pm 2, ~~
289: \gamma = 62_{-12}^{+10}.
290: \eqno{(25)}
291: $$
292: With these we find
293:
294: $$ x_2/J = 2.24 \pm 0.2,
295: \eqno{(26)}
296: $$
297: $$ x_2/x_3 = 4.3_{-1.3}^{+2.7},
298: \eqno{(27)}
299: $$
300: $$ y_3/x_2 = 0.03_{-0.07}^{+0.1},
301: \eqno{(28)}
302: $$
303: and, using $\gamma = \pi -(\alpha+\beta)$ to correlate
304: $\gamma$ and $\alpha$,
305: $$ y_3/x_3 = 0.13_{-0.4}^{+0.2}.
306: \eqno{(29)}
307: $$
308: These results suffer from large uncertainties. However, it is
309: clear that the parameters $x_2, J, x_3, y_3$, are not of the same
310: order. In particular, we find
311: $$ y_3/x_2 = O(\lambda^2).
312: \eqno{(30)}
313: $$
314: In fact, the result $y_3/x_2 = O(\lambda^2)$ can be expressed
315: in a variety of forms when we write $y_3 = Re\Gamma_{321}$ in
316: different parametrizations.
317:
318: In the Wolfenstein parametrization,
319: $$ Re\Gamma_{321} = A^2 \lambda^6 [\rho -(\rho^2+\eta^2)].
320: \eqno{(31)}
321: $$
322: For $V^{(s)}$, we have
323: $$ \begin{array}{rcl}
324: Re\Gamma_{321} &=& Re[s_{13}(c_{12}c_{23}-s_{12}s_{23}
325: s_{13}e^{i\delta})(e^{-i\delta} s_{12}s_{23}
326: -c_{12}c_{23}s_{13})]\\
327: &\cong & s_{13} (c_\delta s_{12}s_{23}-s_{13}).
328: \end{array}
329: \eqno{(32)}
330: $$
331: Using Eq.(10), $|V_{ub}|^2 = x_3-y_3$,
332: $|V_{td}|^2 = x_2-y_3$, it follows that
333: $$ Re\Gamma_{321} = \frac{1}{2} [-|V_{td}|^2 -
334: |V_{ub}|^2 + |V_{us}|^2 |V_{cb}|^2],
335: \eqno{(33)}
336: $$
337: where we have used the constraint (Eq.(7))
338: $(x_2 + x_3 \cong |V_{us}|^2 |V_{cb}|^2$, with $x_1 \cong 1)$.
339: Thus, the condition $y_3/x_2 = O(\lambda^2)$ takes on a variety
340: of forms:
341: $$ 1)~~~-\tan \beta/\tan \alpha = O(\lambda^2);
342: \eqno{(34)}
343: $$
344: $$ 2)~~~[\rho - (\rho^2+\eta^2)]/(1-\rho) = O(\lambda^2);
345: \eqno{(35)}
346: $$
347: $$ 3)~~~\cot^2\delta (c_\delta s_{12}s_{23} -s_{13})/s_{13}
348: = O(\lambda^2);
349: \eqno{(36)}
350: $$
351: $$ 4)~~~\frac{1}{2}(|V_{us}|^2 |V_{cb}|^2 - |V_{td}|^2 -
352: |V_{ub}|^2)/|V_{td}|^2 = O(\lambda^2) .
353: \eqno{(37)}
354: $$
355: Here, in 3), we have used the approximate relations
356: $\delta = \gamma$, $\tan\beta \tan\gamma = 1$ to obtain
357: $x_2 = x_3 \tan^2\delta = s_{13}^2 \tan^2 \delta$.
358: All of these relations are reasonably well satisfied by
359: using the values of the global CKM fits [7]. They are,
360: approximately, $\rho=0.19 \pm 0.08, \eta=0.36 \pm 0.05; s_{12}=0.226 \pm
361: 0.002$, $s_{13} = (3.9 \pm 0.3) \times 10^{-3},
362: s_{23} = (41 \pm 1) \times 10^{-3}$,
363: $\delta = 62 \pm 10$; $|V_{us}| = 0.23 \pm 0.002$,
364: $|V_{ub}| = (3.9 \pm 0.3)\times 10^{-3}$,
365: $|V_{cb}| = (41 \pm 1)\times 10^{-3}$, $|V_{td}| =
366: (8.3 \pm 0.8) \times 10^{-3}$. Eqs.(34-37) exhibit the
367: intriguing correlations amongst the $V_{CKM}$ parameters.
368: Without them $y_3$ would be of the same order as $x_2$.
369:
370: The above analysis can be summarized as an expansion in
371: $\lambda^2$:
372: $$ (-y_2, -y_1, x_2, y_3) = (\lambda^2,A^2\lambda^4, B^2\lambda^6,
373: C\lambda^8),
374: \eqno{(38)}
375: $$
376: with
377: $$ A^2 = 0.64 \pm 0.05
378: \eqno{(39)}
379: $$
380: $$ B^2 = 0.5 \pm 0.1.
381: \eqno{(40)}
382: $$
383: The parameter $C$ is very poorly determined. We use the above
384: relations and make a rough estimate to find
385: $$ C = 0.3 \pm 1.
386: \eqno{(41)}
387: $$
388: Thus, a convenient parametrization of $V_{CKM}$ is to use the
389: set $(y_2,y_1,x_2,y_3)$. It gives rise to an expansion in powers
390: of $\lambda^2$, in contrast to the use of $V^{(W)}$ or
391: $V^{(s)}$, whose matrix elements can be regarded as expansions
392: in powers of $\lambda$. The difference originates from rephasing
393: considerations. For, when one uses rephasing invariants (which
394: are what enter into physical quantities) constructed
395: out of $V^{(W)}$ or $V^{(s)}$, such as $|V_{ij}|^2$ or
396: $V_{\alpha i} V_{\beta j} V_{\alpha j}^* V_{\beta i}^*$ or
397: $(x_i,y_j)$, it is seen that the expansion parameters is
398: $\lambda^2$, and not $\lambda$. Thus, e.g., $|V_{ij}|^2$ have values which
399: are of order $O(1),O(\lambda^2),O(\lambda^4),O(\lambda^6)$. However, the
400: combination $y_3$ (Eq.(33)) contains a cancellation, resulting in
401: $y_3=O(\lambda^8)$. It is this feature which distinguishes the $(x,y)$ set
402: from other parametrizations.
403:
404: The result $y_3 = O(\lambda^8)$ has another interesting consequence.
405: To the extent that all available measurements on $V_{CKM}$ are only
406: accurate to $O(\lambda^6)$, we should be able to set
407: $y_3 = 0$ and fit the extant data on $V_{CKM}$ by three
408: parameters, $(y_2,y_1,x_2)$. This will be presented in Fig.2.
409: The inputs are taken from those of the well-known unitarity triangle
410: construction [7,8,9] in terms of $\rho$ and $\eta$ (since we are
411: only interested in leading order effects, we take
412: $\bar{\rho} = \rho, \bar{\eta} = \eta)$. With
413: $y_2 = -A^2 \lambda^4, y_1 = -\lambda^2, y_3 = 0, x_1 =1 -\lambda^2$,
414: all of the physical quantities $(\epsilon_K,|V_{ub}|^2,
415: |V_{td}|^2, \tan \beta)$ can be converted from functions of
416: $(\rho,\eta)$ into those of $(x_2,x_3)$ using
417: $A^2 \lambda^6 (1-2\rho) = x_2 -x_3$,
418: $A^2\lambda^6 \eta = \sqrt{x_2x_3}$. In addition, we use
419: $J^2 = x_2x_3$ and the constraint $x_2 + x_3 \cong y_1y_2$.
420: The values of the physical quantities are taken from Ref.[9].
421: It is seen that there is a shaded region in the $(x_2,x_3)$ plane where
422: all constraints meet. Fig.2 represents a three parameter
423: $(y_2,y_1,x_2)$ fit (with $x_3 = y_1y_2 -x_2)$ to existing data
424: on $V_{CKM}$. Qualitatively, we can understand the viability of a
425: three parameter fit as follows. The parameter $\eta$, as determined
426: in Ref.[7], has errors of order $O(\lambda^2)$. Setting $y_3=0$,
427: according to Eq.(35), amounts to eliminating $\eta$ and using
428: $\eta=\sqrt{\rho-\rho^2}$. This relation is satisfied to
429: $O(\lambda^2)$. Thus, Fig.2 represents a fit with the parameter
430: set $(\lambda^2,A^2 \lambda^4, \rho,\eta=\sqrt{\rho-\rho^2})$, and
431: should be robust at the $O(\lambda^2)$ level, as demonstrated.
432: Similar results also follow if we use other parametrizations. The above
433: analysis also suggests that whether a non-vanishing $y_3$ is needed has
434: to wait until more precision measurements become available.
435:
436: \begin{figure}[htp]
437: \centering
438: %\epsfxsize=1.3cm
439: %\epsffile{ckm-graph11.eps}
440: \includegraphics[width=12cm]{ckm-graph12.eps}
441: \caption{A three parameter fit to $V_{CKM}$,
442: with fixed $y_1(=-A^2 \lambda^4),
443: y_2(=-\lambda^2)$, and the constraint
444: $x_2+x_3=y_1 y_2$. Values of the physical
445: quantities are taken from Ref.[9].}\label{fig:ckm-graph12.eps}
446: \end{figure}
447:
448: We close with a few concluding remarks. In this paper we argue that,
449: although our present knowledge on $V_{CKM}$ is still far from
450: precise, it nevertheless offers considerable evidence of intriguing
451: correlations, as in Eqs.(34-37). We can take advantage of this
452: by using a particular set of parameters, $(y_2,y_1,x_2,y_3)$,
453: which gives rise to a natural expansion in powers of $\lambda^2$.
454: It is shown that, at the current level of accuracy, we can set
455: $y_3 = O(\lambda^8) \cong 0$ and obtain a three parameter fit to
456: $V_{CKM}$.
457: Within the currently available data set, the proper way to estimate
458: a ``best" value for $y_3$ would be to do a global fit with the
459: $(x,y)$ variables. However, this is beyond the scope of the present
460: paper. Of course, we would have a better handle in the future,
461: when more and better measurements are performed.
462:
463: One important motivation in analyzing $V_{CKM}$ is to check its
464: consistency by over-contraining its parameters, with an eye for
465: discrepancies which might originate from ``new physics". It is
466: our hope that a parametrization which has a distinct hierarchy
467: can help to identify features that will stand out, so that the
468: success or failure of a model can be better assessed.
469:
470: \begin{center}
471: This work is supported in part by DOE grant No.
472: DE-FG02-91ER40681.
473: \end{center}
474:
475:
476:
477:
478:
479:
480:
481:
482:
483:
484:
485:
486: \pagebreak
487:
488: \begin{center}
489: {\bf REFERENCES}
490: \end{center}
491:
492: \begin{description}
493: \item[[1]] L. Wolfenstein, Phys. Rev. Lett. {\bf 51}, 1945 (1983).
494:
495: \item[[2]] T.K. Kuo and Tae-Hun Lee, Phys. Rev. {\bf D71}, 093011 (2005).
496:
497: \item[[3]] C. Jarlskog, Phys. Rev. Lett. {\bf 55}, 1039 (1985); also in $CP$
498: Violation, Edited by C. Jarlskog (World Scientific, Singapore, 1989).
499:
500: \item[[4]] C. Hamzaoui, Phys. Rev. Lett. {\bf 61}, 35 (1988);
501: G.C. Branco and L. Lavoura, Phys. Lett. {\bf B208}, 123 (1988).
502:
503: \item[[5]] Particle Data Group, Phys. Lett. {\bf B952}, 130 (2004);
504: L.L. Chau and W.Y. Keung, Phys. Rev. Lett. {\bf 53}, 1802
505: (1984); H. Harari and M. Leurer, Phyus. Lett. {\bf B181}, 123 (1986).
506:
507: \item[[6]] A.J. Buras, M.E. Lautenbacher and G. Ostermaier, Phys. Rev. {\bf D50},
508: 3433 (1994).
509:
510: \item[[7]] J. Charles et al. [CKMfitter Collaboration], hep-ph/0406184.
511:
512: \item[[8]] M. Bona et al. [UTfit Collaboration], hep-ph/0501199.
513:
514: \item[[9]] See, e.g., A.J. Buras, hep-ph/0505175, and references therein.
515:
516: \end{description}
517:
518:
519: \end{document}
520: