1: \documentstyle[12pt,epsfig]{article}
2: \textheight 9in
3: \textwidth 6in
4: \voffset -0.6in
5: \hoffset -0.3in
6: \def \app{D_{\pi \pi}}
7: \def \dkk{D_{KK}}
8: \def \b{{\cal B}}
9: \def \bea{\begin{eqnarray}}
10: \def \beq{\begin{equation}}
11: \def \bo{B^0}
12: \def \ko{K^0}
13: \def \bs{B_s}
14: \def \obs{\overline{B}_s}
15: \def \bpb{\stackrel{(-)}{B^0}}
16: \def \cn{Collaboration}
17: \def \cpp{C_{\pi \pi}}
18: \def \ckk{C_{KK}}
19: \def \eea{\end{eqnarray}}
20: \def \eeq{\end{equation}}
21: \def \ite{{\it et al.}}
22: \def \kpb{\stackrel{(-)}{K^0}}
23: \def \lpp{\lambda_{\pi \pi}}
24: \def \lkk{\lambda_{KK}}
25: \def \ob{\overline{B}^0}
26: \def \ok{\overline{K}^0}
27: \def \rpp{R_{\pi \pi}}
28: \def \rkk{R_{KK}}
29: \def \rt{r_{\tau}}
30: \def \s{\sqrt{2}}
31: \def \spp{S_{\pi \pi}}
32: \def \skk{S_{KK}}
33: \begin{document}
34:
35: \begin{flushright}
36: TECHNION-PH-2002-14\\
37: EFI 02-68 \\
38: hep-ph/0203158 \\
39: March 2002 \\
40: \end{flushright}
41:
42: \renewcommand{\thesection}{\Roman{section}}
43: \renewcommand{\thetable}{\Roman{table}}
44: \centerline{\bf WEAK PHASE $\gamma$ FROM $\bs(t) \to K^+K^-$}
45: \medskip
46: \centerline{Michael Gronau}
47: \centerline{\it Physics Department, Technion -- Israel Institute of Technology}
48: \centerline{\it 32000 Haifa, Israel}
49: \medskip
50: \centerline{Jonathan L. Rosner}
51: \centerline{\it Enrico Fermi Institute and Department of Physics}
52: \centerline{\it University of Chicago, Chicago, Illinois 60637}
53: \bigskip
54:
55: \begin{quote}
56:
57: We re-examine the time-dependent rates of $\bs(t) \to K^+ K^-$ and
58: $\obs(t) \to K^+ K^-$, including a lifetime difference between neutral $\bs$
59: mass eigenstates. The two rates, normalized by the rate of $\bs \to \ko \ok$,
60: are used to obtain ambiguity-free information on a strong phase and on the
61: weak phase $\gamma$. We discuss the sensitivity of extracting $\gamma$ to
62: the measured quantities, and find that an error of $\pm 10^\circ$ in $\gamma$
63: is possible for a sample of several thousand $\bs(t) \to K^+ K^-$
64: decays. This study is complementary to a recent similar analysis of the U-spin
65: related decays $\bo(t) \to \pi^+\pi^-$ and $\ob(t) \to \pi^+\pi^-$.
66:
67: \end{quote}
68:
69: \leftline{\qquad PACS codes: 12.15.Hh, 12.15.Ji, 13.25.Hw, 14.40.Nd}
70:
71: \section{Introduction}
72:
73: Time-dependent CP asymmetry measurements in neutral $B$ meson decays
74: provide useful tests for the standard model, which attributes CP
75: violation to a single phase in the Cabibbo-Kobayashi-Maskawa (CKM) matrix.
76: The potential variety of $\bo$ and $\bs$ decay modes permits one to
77: overconstrain the CKM unitarity triangle involving the three angles
78: $\alpha = \phi_2,~\beta \equiv \phi_1$ and $\gamma = \phi_3$ \cite{PDG}.
79: The observation of CP violation in $B^0(t) \to J/\psi K_S$ \cite{betaBa,betaBe},
80: interpreted in terms of $\sin 2\beta$, passed this test successfully.
81: First results of a CP asymmetry measurement in $\bo(t) \to \pi^+\pi^-$,
82: consistent with a zero asymmetry, though involving large errors, were reported
83: in \cite{CSBa}.
84: Updates, implying world average asymmetries which are nonzero at about a
85: $2.3\sigma$ level, were presented very recently \cite{Moriond1,Moriond2}.
86: The two measured quantities, coefficients of terms involving
87: $\sin \Delta m t$ and $\cos \Delta m t$ in the time-dependent asymmetry, were
88: studied recently \cite{GR} as functions of relevant strong and weak phases.
89: It was shown how discrete ambiguities in a strong phase can be resolved in
90: order to learn the weak phase $\alpha$.
91:
92: Approximate flavor SU(3) symmetry of strong interactions can be used to
93: relate a variety of $B$ and $\bs$ decays to two charmless pseudoscalar
94: mesons \cite{GHLR}. Thus, the process $\bo \to\pi^+\pi^-$ is related to
95: $\bs \to K^+K^-$ in the limit of U-spin symmetry involving an interchange of
96: $d$ and $s$ quarks. This symmetry implies equal CP rate differences in any
97: pair of U-spin related processes \cite{Uspin}, and in particular in this
98: pair of decay modes. The U-spin relation between
99: $\bo \to\pi^+\pi^-$ and $\bs \to K^+K^-$ was used in \cite{FL} to suggest
100: a simultaneous determination of the two weak phases $\beta$ and $\gamma$ by
101: combining information from the asymmetries in these two processes.
102: By measuring for both channels a $\cos \Delta mt$ and a
103: $\sin\Delta mt$ term one has four equations for four unknowns: $\beta$ (which
104: can also be considered as given from direct measurements), $\gamma$, the ratio
105: of penguin to tree amplitudes in $\bo\to \pi^+\pi^-$, and the relative strong
106: phase between these two amplitudes. It was noted \cite{FL} that factorized
107: U-spin breaking effects cancel in the double ratio of penguin to tree
108: amplitudes and in the difference between the two relative strong phases
109: occuring in the two processes. This reduces the intrinsic theoretical
110: uncertainty of this method.
111:
112: Several earlier suggestions for measuring weak phases in $B_s \to K^+ K^-$
113: were made in \cite{KLY,FL2}, by combining this decay within flavor SU(3)
114: with information from other processes, involving $\bs \to \ko\ok,~\bs
115: \to K^-\pi^+, ~\bo \to K^+\pi^-$ and $B^+ \to \ko \pi^+$. Usually, these
116: methods require some assumption about SU(3) breaking. Also, as a result of
117: the large number of required measurements, which are expressed in terms
118: of trigonometric functions of strong and weak phases, one
119: encounters several discrete ambiguities in the solutions for weak phases.
120:
121: In the present paper we wish to re-examine more carefully the simplest
122: among all these suggestions \cite{KLY}, which does not require full SU(3)
123: but only isospin symmetry. The method is based primarily on the time-dependent
124: rate of $\bs(t) \to K^+K^-$ and its charge-conjugate, normalized by the
125: time-integrated rate of $\bs \to \ko\ok$. We will also include effects due to
126: a lifetime difference between neutral $\bs$ mass eigenstates which were
127: neglected in \cite{KLY}.
128:
129: A major purpose of this study will be to show how to
130: resolve certain discrete ambiguities in strong and weak phases in order to
131: determine unambiguously the weak phase $\gamma$. One of the outputs of the
132: above measurements is the ratio of tree to penguin amplitudes in $\bs
133: \to K^+K^-$. This ratio is related by U-spin to the corresponding ratio in
134: $\bo \to \pi^+\pi^-$, for which some information already exists. We will
135: use this information to demonstrate the sensitivity of this method.
136:
137: We introduce notations for $\bs \to K^+K^-$ and $\bs \to \ko\ok$ in Sec.\ II,
138: where we explain some necessary assumptions and ways of testing these
139: assumptions. The general time-dependence of $\Gamma(\bs(t) \to K^+K^-)$,
140: normalized by $\Gamma(\bs \to \ko\ok)$, involves four measurable quantities
141: $\rkk,~\ckk,~\skk$ and $\dkk$, for which expressions are derived in
142: Sec.\ III in terms of a ratio of tree to penguin amplitude, a strong phase
143: difference and the weak phase $\gamma$. We discuss the resolution of
144: ambiguities associated with the strong phase. Solutions for $\gamma$ in terms
145: of the measurables are studied in Sec.\ IV
146: in order to evaluate the sensitivity of the method. We conclude in Sec.~V.
147:
148:
149: \section{Notation and assumptions}
150:
151: In Ref.\ \cite{GR}, studying $\bo \to \pi^+\pi^-$ in terms of tree and
152: penguin amplitudes,
153: \beq\label{pipi}
154: A(\bo \to \pi^+ \pi^-) = -(|T|e^{i \delta_T} e^{i \gamma} +
155: |P| e^{i \delta_P})~~~,
156: \eeq
157: we used the convention in which top quark contributions are integrated out
158: in the short-distance effective Hamiltonian, and the unitarity relation
159: $V^*_{ub} V_{ud} + V^*_{cb} V_{cd} = - V^*_{tb} V_{td}$ is used. The
160: $V^*_{ub} V_{ud}$ piece of the penguin operator is included in the tree
161: amplitude. We adopt the same convention for $\bs \to K^+ K^-$, such that
162: \beq\label{bs}
163: A(\bs \to K^+ K^-) = -(|T'|e^{i \delta_T'} e^{i \gamma} -
164: |P'| e^{i \delta_P'})~~~.
165: \eeq
166: In the U-spin symmetry limit one has \cite{Uspin, FL}
167: \beq\label{U}
168: \frac{|T'|}{|T|} = \frac{V^*_{ub}V_{us}}{V^*_{ub}V_{ud}} = \frac{\lambda}
169: {1 - \lambda^2/2}~~,~~~~\frac{|P'|}{|P|} =
170: -\frac{V^*_{cb}V_{cs}}{V^*_{cb}V_{cd}} = \frac{1 - \lambda^2/2}{\lambda}~~,
171: ~~~~\delta_T' = \delta_T~~,~~~~\delta_P' = \delta_P~~~,
172: \eeq
173: where $\lambda = 0.22$ \cite{PDG}.
174:
175: We note in passing that Eq.~(\ref{bs}) by itself is a general expression of
176: the decay amplitude in the CKM framework, and involves no assumption.
177: The amplitude for $\obs \to K^+ K^-$ is simply
178: obtained by changing the sign of the weak phase $\gamma$.
179:
180: The amplitude for $\bs \to \ko \ok$ will be assumed to have the simple form
181: \beq\label{bso}
182: A(\bs \to \ko \ok) = -|P'| e^{i \delta_P'}~~~.
183: \eeq
184: The QCD penguin terms in $\bs$ decays to charged and neutral kaons are equal
185: by isospin. However, the form (\ref{bso}) which assumes equal overall penguin
186: contributions in the two processes and no tree amplitude (with weak phase
187: $\gamma$) in $\bs \to \ko\ok$, involves two approximations. First, we neglect a
188: small color-suppressed electroweak penguin contribution in $\bs \to K^+K^-$
189: \cite{GHLR}. Using factorization, such contributions are estimated to be at
190: a level of a few percent \cite{FM}. Tests for the magnitude of the electroweak
191: penguin contribution in $\bs \to K^+ K^-$ were proposed in \cite{FL2}. In
192: Eq.~(\ref{bso}) one also disregards the possibility of enhanced rescattering
193: effects in $\bs \to \ko \ok$, typically of order
194: $|V^*_{ub}V_{us}/V^*_{cb}V_{cs}| \simeq 2\%$ without such
195: enhancement. Such enhanced effects could, in principle, induce a sizable term
196: with phase $\gamma$. One test for such a term would be observing a CP asymmetry
197: between this process and its change conjugate. Other indirect tests involve
198: rate measurements of hadronic decays such as $\bs \to \pi^+\pi^-$ and $\bo
199: \to K^+K^-$, which are very rare in the absence of rescattering \cite{BGR}.
200:
201: Since in the above approximation $\bs \to \ko \ok$ involves a single weak
202: phase, we expect no CP asymmetry in this decay.
203: We denote by $\rkk$ the ratio of charge averaged rates for $\bs~({\rm or}~\obs)
204: \to K^+K^-$ and $\bs~({\rm or}~\obs) \to \ko \ok$,
205: \beq\label{rkk}
206: \rkk \equiv \frac{{\cal B}(\bs \to K^+ K^-) + {\cal B}(\obs \to K^+ K^-)}
207: {{\cal B}(\bs \to \ko \ok) + {\cal B}(\obs \to \ko \ok)}~~~.
208: \eeq
209:
210: The time-dependent decay rate of $\bs|_{\rm initial} \to K^+ K^-$ is given by
211: \cite{isi}
212: \bea\label{Gamma(t)}
213: \Gamma(\bs(t) \to K^+K^-) \propto e^{-\Gamma_s t} \times
214: \left [ \cosh (\Delta\Gamma_s t/2) - \dkk\sinh (\Delta\Gamma_s t/2)
215: \right. \nonumber\\
216: \left.~+ \ckk\cos(\Delta m_s t) -
217: \skk\sin(\Delta m_s t) \right ]~~~,
218: \eea
219: where $\Gamma_s \equiv (\Gamma_H+\Gamma_L)/2$, and $\Delta m_s\equiv m_H -
220: m_L$ and $\Delta\Gamma_s \equiv \Gamma_L - \Gamma_H$ are the mass and width
221: differences between the two $\bs$ mass eigenstates. In the standard model one
222: expects $\Delta\Gamma_s > 0$, and estimates imply that
223: $\Delta\Gamma_s/\Gamma_s$ can be as large as about 15$\%$ \cite{Lenz}.
224: An expression similar to (\ref{Gamma(t)}), in which the
225: $\ckk$ and $\skk$ terms occur with opposite signs, describes the
226: rate of $\obs|_{\rm initial} \to K^+ K^-$.
227:
228: The coefficients $\ckk,~\skk$ and $\dkk$, obeying
229: \beq
230: \ckk^2 + \skk^2 + \dkk^2 = 1~~~,
231: \eeq
232: are defined by
233: \beq \label{CSDkk}
234: \ckk \equiv \frac{1 - |\lkk|^2}{1 + |\lkk|^2}~~,~~~
235: \skk \equiv \frac{2 {\rm Im}(\lkk)}{1 + |\lkk| ^2}~~,~~~
236: \dkk \equiv \frac{2 {\rm Re}(\lkk)}{1 + |\lkk|^2}~~~,
237: \eeq
238: where
239: \beq
240: \lkk \equiv \frac{A(\obs \to K^+ K^-)}{A(\bs \to K^+ K^-)}~~~.
241: \eeq
242: A very small overall phase from $\bs-\obs$ mixing,
243: ${\rm arg}(V^*_{tb}V_{ts}/V^*_{cb}V_{cs})^2 \sim \lambda^2$, was neglected.
244:
245: Defining $\delta' \equiv \delta_P' - \delta_T'$, where we use a convention
246: in which $-\pi \le \delta' \le \pi$, one finds
247: \beq
248: \lkk = \frac{1 - |T'/P'| e^{-i (\delta' + \gamma)}}
249: {1 - |T'/P'| e^{-i (\delta' - \gamma)}}~~~.
250: \eeq
251: This expression simplifies to a pure phase in the two special cases when
252: $\delta'=0$ or $\pi$
253: \beq
254: \lkk = e^{2 i \zeta}~~,~~~~
255: \zeta = \left\{ \begin{array}{c} \arctan \frac{|T'/P'| \sin\gamma}
256: {1 - |T'/P'| \cos\gamma}~~~(\delta' = 0) \cr
257: - \arctan\frac{|T'/P'| \sin\gamma}{1 + |T'/P'| \cos \gamma}~~~(\delta' =
258: \pi)~~~. \end{array} \right.
259: \eeq
260: In such cases $\ckk = 0,~\skk = \sin 2\zeta,~\dkk = \cos 2\zeta$.
261:
262: In the next section we will study the dependence of the measurables defined in
263: Eqs.~(\ref{rkk}) and (\ref{CSDkk}) on $|T'/P'|,~\delta'$ and $\gamma$.
264:
265: \section{Dependence of observables on $\delta'$ and $\gamma$}
266:
267: Expressions for $\rkk,~\ckk,~\skk$ and $\dkk$ in terms of $|T'/P'|,~\delta'$
268: and $\gamma$ are readily derived:
269: \bea
270: \rkk &=& 1 - 2|T'/P'|\cos\delta'\cos\gamma + |T'/P'|^2~~~,\\
271: \ckk &=& -2|T'/P'|\sin\delta'\sin\gamma/\rkk~~~,\\
272: \skk &=& [2|T'/P'|\cos\delta'\sin\gamma - |T'/P'|^2\sin 2\gamma]/\rkk~~~,\\
273: \dkk &=& [1 - 2|T'/P'|\cos\delta'\cos\gamma + |T'/P'|^2\cos 2\gamma]/\rkk~~~.
274: \eea
275:
276: There are some interesting differences between the present case of $B_s
277: \to K^+ K^-$ and that of $B^0 \to \pi^+ \pi^-$ discussed in Ref.\ \cite{GR}.
278:
279: \begin{enumerate}
280:
281: \item Assuming ${\rm sign}(\sin\delta') = {\rm sign}(\sin\delta)$, the
282: quantities $\ckk$ and $\cpp$ are opposite in sign. In the U-spin
283: symmetry limit, the corresponding CP rate differences are equal in
284: magnitude and have opposite signs.
285:
286: \item In contrast to the case of $\spp$,
287: which measures $\sin 2\alpha$ in the limit $|P/T| \to 0$,
288: there is no term independent of the ratio of the smaller amplitude to
289: the larger amplitude; $\skk$ vanishes in the limit $|T'/P'| \to 0$.
290:
291: \item Whereas $\app$ is expected to be large and negative, $\dkk$ is
292: predicted to be large and positive. Its maximal value is +1.
293:
294: \item While one expects $\rpp > 1$ for $\cos \gamma > 0$ and $\cos
295: \delta \simeq 1$, one expects $\rkk < 1$ for $\cos \delta' \simeq 1$.
296:
297: \item All equations involve only the CKM phase $\gamma$, while for
298: $B^0 \to \pi^+ \pi^-$ the relations are not as simple because they
299: contain both $\alpha$ and $\gamma$
300: (or $\beta$ and $\gamma$.) This should not be considered a disadvantage,
301: since $\beta$ is known rather precisely from direct measurements.
302:
303: \end{enumerate}
304:
305: The first three expressions for $\rkk,~\ckk$ and $\skk$ agree with results
306: obtained in \cite{KLY} using different notations. One may regard these three
307: equations as specifying the three unknowns $|T'/P'|,~\delta'$ and $\gamma$.
308: The quadratic equations for $|T'/P'|$ and the trigonometric expressions in
309: terms of strong and weak phases yield solutions which involve several discrete
310: ambiguities.
311:
312: Whereas the magnitude of $\dkk$ is specified by $\ckk$ and $\skk$, which are
313: likely to be more easily measured, its sign could in principle resolve some of
314: these ambiguities. $\dkk$ may change sign as a function of $\delta'$ and
315: $\gamma$ if $|T'/P'|$ is sufficiently large. For instance, in \cite{KLY} this
316: ratio was taken to be $8/9$, which would permit a change of sign in $\dkk$.
317: As we will show now, there exists already indirect information on $|T'/P'|$
318: from nonstrange $B$ decays which implies $|T'/P'| = 0.184 \pm 0.043$,
319: which is much smaller than assumed in \cite{KLY}. With such a small value of
320: $|T'/P'|$, $\dkk$ must be positive and large
321: in the CKM framework. A negative value would signify new physics.
322:
323: Using Eq.\ (\ref{U}), we find to leading order in $\lambda^2$,
324: \beq\label{T'/P'}
325: \frac{|T'|}{|P'|} = \frac{\lambda^2}{1 - \lambda^2}\frac{|T|}{|P|}~~~.
326: \eeq
327: This relation, which is precise in the U-spin symmetry limit, may be affected
328: by U-spin breaking. However, as noted in \cite{FL}, such effects are expected
329: to be small in the ratio of ratios $|T'/P'|:|T/P|$ if one assumes approximate
330: factorization. We will neglect nonfactorizable corrections in our evaluation of
331: $|T'/P'|$. The ratio $|P/T|$ was estimated in Refs.\ \cite{GR01} and
332: \cite{LR}, applying SU(3) with SU(3) breaking to experimental data on
333: $B^+ \to K^0 \pi^+$ (a process dominated by the penguin amplitude), and
334: applying data on $B \to \pi l \nu$ related to the tree amplitude in
335: $\bo \to \pi^+\pi^-$ in the factorization approximation. We shall use the
336: result of Ref.\ \cite{GR01},
337: $|P/T| = 0.276 \pm 0.064$. Ref.\ \cite{BBNS}, based on explicit
338: calculations in QCD-improved factorization, found a very similar value
339: $|P/T| = 0.285 \pm 0.076$. Thus, we find from Eq.~(\ref{T'/P'})
340: \beq\label{T'/P'number}
341: \frac{|T'|}{|P'|} = 0.184 \pm 0.043~~~.
342: \eeq
343: As mentioned, this value allows only positive values of $\dkk$,
344: which cannot provide new information for reducing discrete ambiguities.
345:
346: Let us discuss ambiguities related to the strong phase $\delta'$, which
347: can be positive or negative, and can lie either in the range $0 < |\delta'|
348: \le \pi/2$ or in $\pi/2 < |\delta'| \le \pi$. Calculations based on
349: QCD-improved factorization \cite{BBNS} may indicate that this phase (related
350: by U-spin to the phase $\delta \equiv \delta_P - \delta_T$ defined in
351: (\ref{pipi})) is small and has a definite sign; a value $\delta \simeq
352: 10^\circ$ is found in Ref.\ \cite{BBNS}. Other
353: perturbative QCD calculations \cite{KLS} imply a larger negative phase
354: $\delta$; correspondingly, Ref.\ \cite{Chen} finds $\delta' \simeq
355: -27^\circ$ when the phase is expressed in our convention. It will be
356: important to check these predictions.
357:
358: In order to resolve ambiguities in $\delta'$, we note that $\rkk,~\skk$ and
359: $\dkk$ are even in $\delta'$ while $\ckk$ is odd in $\delta'$. First, consider
360: $\rkk$. Since current constraints on CKM parameters \cite{Beneke,Hocker,Ciu}
361: imply $\gamma <~\pi/2$ or $\cos\gamma > 0$, a value $\rkk < 1$ would imply
362: $\cos\delta' > 0$, namely $0 < |\delta'| < \pi/2$. Furthermore, a value of
363: $\rkk$ below 1 permits one to set a significant bound on $\gamma$ which is
364: independent of $\delta'$ \cite{FM,GR},
365: \beq
366: \sin^2\gamma \le \rkk~~~.
367: \eeq
368:
369: Second, consider $\ckk$. Within the CKM framework $\sin\gamma > 0$. Therefore,
370: a nonzero measurement of $\ckk$ will specify the sign of $\delta'$.
371: There exists an absolute bound on $\ckk$, $|\ckk| \le 2|T'/P'|/(1 +
372: |T'/P'|^2)$, which becomes stronger for a {\it given} value of $\gamma$:
373: \beq
374: |\ckk| \le \frac{2|T'/P'| |\sin\gamma|}{\sqrt{(1 + |T'/P'|^2)^2
375: - 4|T'/P'|^2 \cos^2\gamma}}~~~.
376: \eeq
377:
378: Examining $\skk$, we note that its dominant term is $2|T'/P'|\cos\delta'\sin
379: \gamma$. Hence, neglecting the term quadratic in $|T'/P'|$, $\skk \ge 0$ for $0
380: < |\delta'| \le \pi/2$, and $\skk < 0$ for $\pi/2 < |\delta'| \le \pi$, which
381: resolves the same ambiguity in $\delta'$ as in $\rkk$. In this approximation,
382: $\rkk\skk$ is bound by $|\rkk\skk| \le 2|T'/P'||\sin\gamma|$ for a given
383: value of $\gamma$.
384:
385: \section{Solutions for $\gamma$}
386:
387: Combining the equations for $\rkk$ and $\ckk$, one can eliminate the strong
388: phase $\delta'$
389: \beq\label{rkk1}
390: \rkk = 1 + |T'/P'|^2 \pm \sqrt{4|T'/P'|^2\cos^2\gamma -
391: (\rkk\ckk)^2\cot^2\gamma}~~~.
392: \eeq
393: This equation can be inverted to obtain a quadratic equation for $\sin^2\gamma$
394: in terms of $\rkk$ and $\ckk$ \cite{similar}:
395: \bea
396: & &4|T'/P'|\sin\gamma = \nonumber \\
397: & &\pm \{[(1 + |T'/P'|)^2 - \rkk(1 + \ckk)][\rkk(1 - \ckk) -
398: (1 -|T'/P'|)^2] \}^{1/2} \nonumber\\
399: & &\pm \{[(1 + |T'/P'|)^2 - \rkk(1 - \ckk)][\rkk(1 + \ckk) -
400: (1 - |T'/P'|)^2] \}^{1/2}.
401: \eea
402: As mentioned, only positive solutions of $\sin\gamma$ apply within the CKM
403: framework.
404:
405: While the value of $|T'/P'|$ in Eq.~(\ref{T'/P'number}) was obtained indirectly
406: from nonstrange $B$ decays, a direct cross check in $\bs \to K \bar K$ can
407: be achieved by using information from $\skk$. One obtains
408: \beq\label{rkk2}
409: \rkk(1 + \skk\cot\gamma) = 1- |T'/P'|^2\cos 2\gamma~~~.
410: \eeq
411: Alternatively, neglecting the term in $\skk$ quadratic in $|T'/P'|$, one
412: gets a useful approximation,
413: \beq
414: 4|T'/P'|^2 \sin^2\gamma \approx \rkk^2(\ckk^2 + \skk^2)~~~.
415: \eeq
416:
417: \begin{figure}[t]
418: \centerline{\epsfysize = 4.9 in \epsffile{rkkg18.ps}}
419: \caption{Dependence of $\rkk$ on $\gamma$ for fixed values of $|\rkk\ckk|$
420: (labels on solid curves). Dot-dashed curves correspond to
421: $\ckk = 0$. Dashed curves (labeled by values of $\rkk \skk$)
422: show values of $\rkk$ for fixed values of $\rkk \skk$. Only the
423: range $0 \le \gamma \le 90^{\circ}$ is shown since all equations are
424: invariant under $\gamma \to \pi - \gamma$, $\skk \to - \skk$. These curves
425: were plotted for $|T'/P'| = 0.184$.}
426: \end{figure}
427:
428: Graphical solutions of Eqs.~(\ref{rkk1}) and (\ref{rkk2}) can be used to
429: demonstrate the sensitivity of the extracted value of $\gamma$ to
430: the measured quantities. In Fig.~1 we use these two equations
431: to plot $\rkk$ as a function of $\gamma$ for several values of $\rkk\ckk$ and
432: for several values of $\rkk\skk$. For $|T'/P'|$ we adopt the
433: central value in Eq.~(\ref{T'/P'number}), $|T'/P'| = 0.184$.
434: Figs.~2 and 3 show the effects of varying $|T'/P'|$ by $\pm 1 \sigma$.
435:
436: The curves in Figs.\ 1--3 indicate that $\gamma$ can be specified by
437: measurements of $|T'/P'|$, $\rkk$, $\skk$, and $\ckk$. The equations are
438: overconstrained, allowing the elimination of some discrete ambiguities.
439: Usually, measurements of $\rkk$ and $\ckk$ are seen to leave a twofold
440: ambiguity in the solution for $\gamma$ within the physical CKM range, $0 <
441: \gamma < \pi/2$, and another twofold ambiguity outside this range.
442: This fourfold ambiguity is resolved by $\skk$.
443:
444: \begin{figure}
445: \centerline{\epsfysize = 4.9 in \epsffile{rkkg14.ps}}
446: \caption{Dependence of $\rkk$ on $\gamma$ for fixed values of $|\rkk\ckk|$
447: (solid curves) and $\rkk \skk$ (dashed curves). Same as Fig.\ 1 except
448: $|T'/P'| = 0.141$.}
449: \end{figure}
450:
451: As an example, assume that $|T'/P'| = 0.184$, and
452: let us imagine that we observe $\rkk = 0.8$ and $|\rkk \ckk|
453: = 0.1$. Acceptable solutions are $\gamma \simeq \gamma_1 \equiv 22^\circ$,
454: $\gamma_2 \equiv 47^\circ$, $\pi - \gamma_1$, and $\pi - \gamma_2$. These
455: values correspond, respectively, to $\rkk \skk \simeq 0.07,~0.22,-0.07,-0.22$.
456: Thus, even given perfect measurements of (say) $\rkk$ and $\ckk$, one will
457: need to determine $\rkk \skk$ to an accuracy of $\pm 0.05$ or better in
458: order either to resolve the discrete ambiguity or to encounter an
459: inconsistency.
460: For $|T'/P'| = 0.141$ these measurements of $\rkk$ and $\ckk$ lead to no
461: solution of $\gamma$, and for $|T'/P'| = 0.227$ the ambiguity in $\gamma$
462: can be resolved with a less accurate measurement of $\skk$.
463: This indicates that at least several hundred flavor tagged $B_s \to K^+ K^-$
464: decays would be needed for the present method to be useful.
465: This estimate ignores background and resolution factors.
466: A less optimistic estimate is based on extrapolation of the BaBar
467: result for $B^0 \to \pi^+ \pi^-$ \cite{CSBa}, in which a sample of 65
468: flavor tagged
469: events led to errors of about $\pm (0.5,0.6)$ in $(\cpp,\spp)$. If these
470: errors scale as the inverse square root of the number of events, it may be
471: necessary to obtain several thousand tagged
472: $B_s \to K^+ K^-$ decays to achieve errors of $\pm 0.05$ in $\ckk$ and $\skk$.
473:
474: \begin{figure}
475: \centerline{\epsfysize = 4.9 in \epsffile{rkkg23.ps}}
476: \caption{Dependence of $\rkk$ on $\gamma$ for fixed values of $|\rkk\ckk|$
477: (solid curves) and $\rkk \skk$ (dashed curves). Same as Fig.\ 1 except
478: $|T'/P'| = 0.227$.}
479: \end{figure}
480:
481: Equations (\ref{rkk1}) and (\ref{rkk2}) can be used to obtain both $|T'/P'|$
482: and $\gamma$. Eliminating $|T'/P'|$,
483: {\em one expresses $\gamma$ in terms of measurable quantities},
484: % \newpage
485: \bea\label{RS}
486: \rkk\skk & = & -\frac{1}{2}\sin 4\gamma - (\rkk - 1)\sin 2\gamma
487: \nonumber\\
488: & \pm & \cos 2\gamma \sqrt{\sin^2 2\gamma + 4(\rkk - 1)\sin^2\gamma -
489: (\rkk\ckk)^2}~~~,
490: \eea
491: while $|T'/P'|$ can be obtained from (\ref{rkk2}).
492: Eq.~(\ref{RS}) provides the essence of this method of determining $\gamma$.
493: Although the $\pm$ sign in this equation seems to indicate a possible
494: discrete ambiguity in the solution for $\gamma$, we have shown in the above
495: numerical examples that the three observables $\rkk, \ckk$ and $\skk$
496: determine the weak phase in a unique manner. This follows from the
497: requirement that $|T'/P'|$ lies in the range (\ref{T'/P'number}).
498:
499: Since the direct asymmetry $\ckk$ occurs quadratically in Eq.~(\ref{RS}),
500: $\gamma$ is relatively insensitive to small values of this observable.
501: Small values correspond to small values of $\delta'$ as predicted in
502: \cite{BBNS}. The quantity $\rkk$, defined in (\ref{rkk}), is the
503: ratio of two charged averaged rates, the measurements of which do not require
504: flavor tagging of neutral $\bs$ mesons. The error in this quantity is expected
505: to be much smaller than in the mixing induced asymmetry $\skk$, which requires
506: both flavor tagging and time-dependence. Therefore, for small values of
507: $\rkk\ckk$, the dominant uncertainty in $\gamma$ originates in the error of
508: $\skk$.
509:
510: As an illustrative example, suppose $\rkk = 0.85 \pm 0.05$, $\rkk \ckk \le
511: 0.1$, and $\rkk \skk = 0.25 \pm 0.05$. We find that
512: $\gamma$ is indeed insensitive to $\rkk \ckk$. The
513: lowest value $\gamma = 50^\circ$ is attained when $\rkk = 0.8$ and $\rkk
514: \skk = 0.25$, while the highest value $\gamma = 70^\circ$ is attained when
515: $\rkk = 0.9$ and $\skk = 0.35$. It can be checked that the values of $|T'/P'|$
516: for these solutions are within the allowed range. Thus, an error on $\gamma$
517: of $\pm 10^\circ$, following from an error of $\pm 0.05$ in $\rkk\skk$,
518: requires a sample of several hundred $B_s \to K^+ K^-$ decays.
519: A realistic number of required flavor tagged events,
520: including background and resolution factors, is more like a few thousand.
521: Plans to measure $\bs(t) \to K^+K^-$ at the Tevatron Run II, with a comparable
522: precision in $\skk$, are described in Ref.~\cite{Bworkshop}.
523:
524: \section{Conclusions}
525:
526: We have discussed time-dependent observables in the decay $B_s \to K^+
527: K^-$ in a manner parallel to that \cite{GR} for $B^0 \to \pi^+ \pi^-$.
528: Measurements based on the $B_s$ decay alone (when combined with a
529: measurement of the rate for $B_s \to K^0 \ok$) suffice to specify
530: both the weak phase $\gamma$ and the strong phase $\delta'$ which
531: is the difference between the penguin and tree strong phases. The
532: measurements are capable of resolving several discrete ambiguities. In order
533: to achieve an error on $\gamma$ of $\pm 10^\circ$, it will be necessary to
534: obtain a sample of several thousand flavor tagged $B_s \to K^+ K^-$ decays.
535:
536: This method does not require full flavor SU(3) or discrete U-spin but only
537: isospin symmetry.
538: We neglected two small contributions, a color-suppressed penguin amplitude in
539: $\bs \to K^+K^-$, and rescattering effects in $\bs \to \ko \ok$, as well
540: as a small phase in $\bs-\obs$ mixing. Each of these terms is expected to
541: introduce an uncertainty in $\gamma$ of only a few percent.
542:
543: The observable $\dkk$, whose measurement depends on the observability of the
544: width difference between the strange $B$ mass eigenstates, is predicted to
545: be large and positive. (Its maximum value is $+1$.)
546: Any deviation from this prediction would be evidence
547: for physics beyond the standard Kobayashi-Maskawa picture of CP violation.
548:
549: \section*{Acknowledgments}
550:
551: This work was partially supported by the US - Israel Binational Science
552: Foundation through Grant No. 98-00237.
553: The research of J. L. R. was supported in part by the United States Department
554: of Energy through Grant No.\ DE FG02 90ER40560, while that of M. G. -- by
555: the Israel Science Foundation founded by the Israel Academy of Sciences and
556: Humanities.
557:
558: % Journal definitions
559: \def \ajp#1#2#3{Am.\ J. Phys.\ {\bf#1}, #2 (#3)}
560: \def \apny#1#2#3{Ann.\ Phys.\ (N.Y.) {\bf#1}, #2 (#3)}
561: \def \app#1#2#3{Acta Phys.\ Polonica {\bf#1}, #2 (#3)}
562: \def \arnps#1#2#3{Ann.\ Rev.\ Nucl.\ Part.\ Sci.\ {\bf#1}, #2 (#3)}
563: \def \art{and references therein}
564: \def \cmts#1#2#3{Comments on Nucl.\ Part.\ Phys.\ {\bf#1}, #2 (#3)}
565: \def \cn{Collaboration}
566: \def \cp89{{\it CP Violation,} edited by C. Jarlskog (World Scientific,
567: Singapore, 1989)}
568: \def \econf#1#2#3{Electronic Conference Proceedings {\bf#1}, #2 (#3)}
569: \def \efi{Enrico Fermi Institute Report No.}
570: \def \epjc#1#2#3{Eur.\ Phys.\ J.\ C {\bf#1}, #2 (#3)}
571: \def \f79{{\it Proceedings of the 1979 International Symposium on Lepton and
572: Photon Interactions at High Energies,} Fermilab, August 23-29, 1979, ed. by
573: T. B. W. Kirk and H. D. I. Abarbanel (Fermi National Accelerator Laboratory,
574: Batavia, IL, 1979}
575: \def \hb87{{\it Proceeding of the 1987 International Symposium on Lepton and
576: Photon Interactions at High Energies,} Hamburg, 1987, ed. by W. Bartel
577: and R. R\"uckl (Nucl.\ Phys.\ B, Proc.\ Suppl., vol. 3) (North-Holland,
578: Amsterdam, 1988)}
579: \def \ib{{\it ibid.}~}
580: \def \ibj#1#2#3{~{\bf#1}, #2 (#3)}
581: \def \ichep72{{\it Proceedings of the XVI International Conference on High
582: Energy Physics}, Chicago and Batavia, Illinois, Sept. 6 -- 13, 1972,
583: edited by J. D. Jackson, A. Roberts, and R. Donaldson (Fermilab, Batavia,
584: IL, 1972)}
585: \def \ijmpa#1#2#3{Int.\ J.\ Mod.\ Phys.\ A {\bf#1}, #2 (#3)}
586: \def \ite{{\it et al.}}
587: \def \jhep#1#2#3{JHEP {\bf#1}, #2 (#3)}
588: \def \jpb#1#2#3{J.\ Phys.\ B {\bf#1}, #2 (#3)}
589: \def \lg{{\it Proceedings of the XIXth International Symposium on
590: Lepton and Photon Interactions,} Stanford, California, August 9--14, 1999,
591: edited by J. Jaros and M. Peskin (World Scientific, Singapore, 2000)}
592: \def \lkl87{{\it Selected Topics in Electroweak Interactions} (Proceedings of
593: the Second Lake Louise Institute on New Frontiers in Particle Physics, 15 --
594: 21 February, 1987), edited by J. M. Cameron \ite~(World Scientific, Singapore,
595: 1987)}
596: \def \kdvs#1#2#3{{Kong.\ Danske Vid.\ Selsk., Matt-fys.\ Medd.} {\bf #1}, No.\
597: #2 (#3)}
598: \def \ky{{\it Proceedings of the International Symposium on Lepton and
599: Photon Interactions at High Energy,} Kyoto, Aug.~19-24, 1985, edited by M.
600: Konuma and K. Takahashi (Kyoto Univ., Kyoto, 1985)}
601: \def \mpla#1#2#3{Mod.\ Phys.\ Lett.\ A {\bf#1}, #2 (#3)}
602: \def \nat#1#2#3{Nature {\bf#1}, #2 (#3)}
603: \def \nc#1#2#3{Nuovo Cim.\ {\bf#1}, #2 (#3)}
604: \def \nima#1#2#3{Nucl.\ Instr.\ Meth.\ A {\bf#1}, #2 (#3)}
605: \def \np#1#2#3{Nucl.\ Phys.\ {\bf#1}, #2 (#3)}
606: \def \npps#1#2#3{Nucl.\ Phys.\ Proc.\ Suppl.\ {\bf#1}, #2 (#3)}
607: \def \os{XXX International Conference on High Energy Physics, Osaka, Japan,
608: July 27 -- August 2, 2000}
609: \def \PDG{Particle Data Group, D. E. Groom \ite, \epjc{15}{1}{2000}}
610: \def \pisma#1#2#3#4{Pis'ma Zh.\ Eksp.\ Teor.\ Fiz.\ {\bf#1}, #2 (#3) [JETP
611: Lett.\ {\bf#1}, #4 (#3)]}
612: \def \pl#1#2#3{Phys.\ Lett.\ {\bf#1}, #2 (#3)}
613: \def \pla#1#2#3{Phys.\ Lett.\ A {\bf#1}, #2 (#3)}
614: \def \plb#1#2#3{Phys.\ Lett.\ B {\bf#1}, #2 (#3)}
615: \def \pr#1#2#3{Phys.\ Rev.\ {\bf#1}, #2 (#3)}
616: \def \prc#1#2#3{Phys.\ Rev.\ C {\bf#1}, #2 (#3)}
617: \def \prd#1#2#3{Phys.\ Rev.\ D {\bf#1}, #2 (#3)}
618: \def \prl#1#2#3{Phys.\ Rev.\ Lett.\ {\bf#1}, #2 (#3)}
619: \def \prp#1#2#3{Phys.\ Rep.\ {\bf#1}, #2 (#3)}
620: \def \ptp#1#2#3{Prog.\ Theor.\ Phys.\ {\bf#1}, #2 (#3)}
621: \def \rmp#1#2#3{Rev.\ Mod.\ Phys.\ {\bf#1}, #2 (#3)}
622: \def \rp#1{~~~~~\ldots\ldots{\rm rp~}{#1}~~~~~}
623: \def \zpd#1#2#3{Zeit.\ Phys.\ D {\bf#1}, #2 (#3)}
624:
625: \begin{thebibliography}{99}
626:
627: \bibitem{PDG} \PDG.
628:
629: \bibitem{betaBa} BaBar \cn, B. Aubert \ite, \prl{87}{091801}{2001};
630: hep-ex/0203007, BABAR-CONF-02/01, presented at the 16th Rencontres de
631: Physique de la Vallee d'Aoste, La Thuile, Italy, March 3$-$9, 2002.
632:
633: \bibitem{betaBe} Belle \cn, K. Abe \ite, \prl{87}{091802}{2001};\\
634: hep-ex/0202027, submitted to Phys. Rev. D; talk presented by K. Trabelsi
635: at XXXVIIth Rencontres de Moriond on Electroweak Interactions and Unified \\
636: Theories, Les Arcs, France, March 9--16, 2002.
637:
638: \bibitem{CSBa} BaBar \cn, B. Aubert \ite, \prd{65}{051502(R)}{2002}.
639:
640: \bibitem{Moriond1} Belle \cn, talk by K. Trabelsi at XXXVIIth Rencontres de
641: Moriond on Electroweak Interactions and Unified Theories, Les Arcs, France,
642: March 9$-$16, 2002.
643:
644: \bibitem{Moriond2} BaBar \cn, talk by A. Farbin at XXXVIIth Rencontres de
645: Moriond on Electroweak Interactions and Unified Theories, Les Arcs, France,
646: March 9$-$16, 2002.
647:
648: \bibitem{GR} M. Gronau and J. L. Rosner, hep-ph/0202170,
649: to be published in Phys. Rev. D.
650:
651: \bibitem{GHLR} M. Gronau, O. F. Hern\'andez, D. London, and J. L. Rosner,
652: \prd{50}{4529}{1994}; \prd{52}{6374}{1995}.
653:
654: \bibitem{Uspin} M. Gronau, \plb{492}{297}{2000}.
655:
656: \bibitem{FL} R. Fleischer, \plb{459}{306}{1999}.
657:
658: \bibitem{KLY} C. S. Kim, D. London and T. Yoshikawa, \prd{57}{4010}{1998}.
659:
660: \bibitem{FL2} R. Fleischer, \prd{58}{093001}{1998}.
661:
662: \bibitem{FM} R. Fleischer and T. Mannel, \prd{57}{2752}{1998}.
663:
664: \bibitem{BGR} B. Blok, M. Gronau and J. L. Rosner, Phys. Rev. Lett.
665: {\bf78} (1997) 3999; {\bf 79} (1997) 1167;
666: M. Gronau and J. L. Rosner, \prd{58}{113005}{1998}.
667:
668: \bibitem{isi} I. Dunietz, \prd{52}{3048}{1995}.
669:
670: \bibitem{Lenz} Talks given by A. Lenz and L. Lellouch at the Workshop on the
671: CKM Unitarity Triangle, CERN, Geneva, February 13$-$16, 2002.
672:
673: \bibitem{GR01} M. Gronau and J. L. Rosner, \prd{65}{013004}{2002}.
674:
675: \bibitem{LR} Z. Luo and J. L. Rosner, \prd{65}{054027}{2002}.
676:
677: \bibitem{BBNS} M. Beneke, G. Buchalla, M. Neubert, and C. T. Sachrajda,
678: \np{B606}{245}{2001}.
679:
680: \bibitem{KLS} Y. Y. Keum and H. N. Li, \prd{63}{074006}{2001}.
681:
682: \bibitem{Chen} C.-H. Chen, \plb{520}{33}{2001}.
683:
684: \bibitem{Beneke} M. Beneke, RWTH Aachen preprint PITHA 02/03,
685: hep-ph/0201337, talk presented at the International Europhysics
686: Conference on High Energy Physics, Budapest, July 2001.
687:
688: \bibitem{Hocker} A. H\"ocker, H. Lacker, S. Laplace, and F. LeDiberder,
689: \epjc{21}{225}{2001}, updated in hep-ph/0112295 (unpublished), presented
690: at Heavy Flavors 9 Conference, Pasadena, CA, September 2001.
691:
692: \bibitem{Ciu} M. Ciuchini \ite, \jhep{0107}{013}{2001}.
693:
694: \bibitem{similar} Analogous expressions for $\sin \gamma$ were obtained
695: for $B \to DK$ by M. Gronau and D. Wyler, \plb{265}{172}{1991}, and for
696: $B \to K\pi$ by M. Gronau and J. L. Rosner, \prd{57}{6843}{1998}.
697:
698: \bibitem{Bworkshop} K. Anikeev \ite, Workshop on B Physics at the Tevatron:
699: Run II and Beyond, Batavia, Illinois, 23$-$25 Sep 1999, and 24$-$26 Feb 2000,
700: Fermilab Report No. FERMILAB-PUB-01-197, hep-ph/0201071.
701:
702: \end{thebibliography}
703: \end{document}
704: #!/bin/csh -f
705: # this uuencoded Z-compressed .tar file created by csh script uufiles
706: # for more information, see e.g. http://xxx.lanl.gov/faq/uufaq.html
707: # if you are on a unix machine this file will unpack itself:
708: # strip off any mail header and call resulting file, e.g., bskkfigs.uu
709: # (uudecode ignores these header lines and starts at begin line below)
710: # then say csh bskkfigs.uu
711: # or explicitly execute the commands (generally more secure):
712: # uudecode bskkfigs.uu ; uncompress bskkfigs.tar.Z ;
713: # tar -xvf bskkfigs.tar
714: # on some non-unix (e.g. VAX/VMS), first use an editor to change the
715: # filename in "begin" line below to bskkfigs.tar_Z , then execute
716: # uudecode bskkfigs.uu
717: # compress -d bskkfigs.tar_Z
718: # tar -xvf bskkfigs.tar
719: #
720: uudecode $0
721: chmod 644 bskkfigs.tar.Z
722: zcat bskkfigs.tar.Z | tar -xvf -
723: rm $0 bskkfigs.tar.Z
724: exit
725:
726: