1: \documentstyle[pra,aps,epsfig]{revtex}
2: %\documentstyle[pra,aps,preprint]{revtex}
3: \begin{document}
4: \draft
5: \title{Real measurements and Quantum Zeno effect}
6: \author{Julius Ruseckas and B. Kaulakys}
7: \address{Institute of Theoretical Physics and Astronomy,\\
8: A. Go\v{s}tauto 12, 2600 Vilnius, Lithuania}
9: \date{\today}
10: \maketitle
11:
12: \begin{abstract}
13: In 1977, Mishra and Sudarshan showed that an unstable particle would never be
14: found decayed while it was continuously observed. They called this effect the
15: quantum Zeno effect (or paradox). Later it was realized that the frequent
16: measurements could also accelerate the decay (quantum anti-Zeno effect). In
17: this paper we investigate the quantum Zeno effect using the definite model of
18: the measurement. We take into account the finite duration and the finite
19: accuracy of the measurement. A general equation for the jump probability during
20: the measurement is derived. We find that the measurements can cause inhibition
21: (quantum Zeno effect) or acceleration (quantum anti-Zeno effect) of the
22: evolution, depending on the strength of the interaction with the measuring
23: device and on the properties of the system. However, the evolution cannot be
24: fully stopped.
25: \end{abstract}
26: \pacs{03.65.Xp, 03.65.Ta}
27:
28: \section{Introduction}
29:
30: Theory of measurements has a special status in quantum mechanics. Unlike
31: classical mechanics, in quantum mechanics it cannot be assumed that the effect
32: of the measurement on the system can be made arbitrarily small. It is necessary
33: to supplement quantum theory with additional postulates, describing the
34: measurement. One of such additional postulate is von Neumann's state reduction
35: (or projection) postulate \cite{vNeum}. The essential peculiarity of this
36: postulate is its nonunitary character. However, this postulate refers only to
37: an ideal measurement, which is instantaneous and arbitrarily accurate. Real
38: measurements are described by the projection postulate only roughly.
39:
40: The important consequence of von Neumann's projection postulate is the quantum
41: Zeno effect. In quantum mechanics short-time behavior of nondecay probability
42: of unstable particle is not exponential but quadratic \cite{Khalfin}. This
43: deviation from the exponential decay has been observed by Wilkinson {\it et
44: al.\/} \cite{GG}. In 1977, Mishra and Sudarshan \cite{Mishra} showed that this
45: behavior when combined with the quantum theory of measurement, based on the
46: assumption of the collapse of the wave function, leaded to a very surprising
47: conclusion: frequent observations slowed down the decay. An unstable particle
48: would never decay when continuously observed. Mishra and Sudarshan have called
49: this effect the quantum Zeno paradox or effect. The effect is called so in
50: allusion to the paradox stated by Greek philosopher Zeno (or Zenon) of Elea.
51: The very first analysis does not take into account the actual mechanism of the
52: measurement process involved, but it is based on an alternating sequence of
53: unitary evolution and a collapse of the wave function. The Zeno effect has been
54: experimentally proved \cite{Itano} in a repeatedly measured two-level system
55: undergoing Rabi oscillations. The outcome of this experiment has also been
56: explained without the collapse hypothesis \cite{Petrosky,FS,Namiki}.
57:
58: Later it was realized that the repeated measurements could not only slow the
59: quantum dynamics but the quantum process may be accelerated by frequent
60: measurements as well
61: \cite{Kofman1,Kaulakys,Pascazio2,Elattari,Facchi,Kofman2,Lewenstein}. This
62: effect was called a quantum anti-Zeno effect by Kaulakys and Gontis
63: \cite{Kaulakys}, who argued that frequent interrogations may destroy quantum
64: localization effect in chaotic systems. An effect, analogous to the quantum
65: anti-Zeno effect has been obtained in a computational study involving barrier
66: penetration, too \cite{Fearn}. Recently, an analysis of the acceleration of a
67: chemical reaction due to the quantum anti-Zeno effect has been presented in
68: Ref. \cite{prezhdo}.
69:
70: Although a great progress in the investigation of the quantum Zeno effect has
71: been made, this effect is not completely understood as yet. In the analysis of
72: the quantum Zeno effect the finite duration of the measurement becomes
73: important, therefore, the projection postulate is not sufficient to solve this
74: problem. The complete analysis of the Zeno effect requires a more precise model
75: of measurement than the projection postulate.
76:
77: The purpose of this article is to consider such a model of the measurement. The
78: model describes a measurement of the finite duration and finite accuracy.
79: Although the used model does not describe the irreversible process, it leads,
80: however, to the correct correlation between the states of the measured system
81: and the measuring apparatus.
82:
83: Due to the finite duration of the measurement it is impossible to consider
84: infinitely frequent measurements, as in Ref. \cite{Mishra}. The highest
85: frequency of the measurements is achieved when the measurements are performed
86: one after another, without the period of the measurement-free evolution between
87: two successive measurements. In this paper we consider such a sequence of the
88: measurements. Our goal is to check whether this sequence of the measurements
89: can change the evolution of the system and to verify the predictions of the
90: quantum Zeno effect.
91:
92: The work is organized as follows. In section \ref{sec:mod} we present the model
93: of the measurement. A simple case is considered in section \ref{sec:id} in
94: order to determine the requirements for the duration of the measurement. In
95: section \ref{sec:meas} we derived a general formula for the probability of the
96: jump into another level during the measurement. The effect of repeated
97: measurements on the system with a discrete spectrum is investigated in section
98: \ref{sec:discr}. The decaying system is considered in section \ref{sec:dec}.
99: Section \ref{sec:concl} summarizes our findings.
100:
101: \section{Model of the measurements}
102: \label{sec:mod}
103:
104: We consider a system which consists of two parts. The first part of the system
105: has the discrete energy spectrum. The Hamiltonian of this part is $\hat{H}_0$.
106: The other part of the system is represented by Hamiltonian $\hat{H}_1$.
107: Hamiltonian $\hat{H}_1$ commutes with $\hat{H}_0$. In a particular case the
108: second part can be absent and $\hat{H}_1$ can be zero. The operator
109: $\hat{V}(t)$ causes the jumps between different energy levels of $\hat{H}_0$.
110: Therefore, the full Hamiltonian of the system equals to
111: $\hat{H}_S=\hat{H}_0+\hat{H}_1+\hat{V}(t)$. The example of such a system is an
112: atom with the Hamiltonian $\hat{H}_0$ interacting with the electromagnetic
113: field, represented by $\hat{H}_1$.
114:
115: We will measure in which eigenstate of the Hamiltonian $\hat{H}_0$ the
116: system is. The measurement is performed by coupling the system with the
117: detector. The full Hamiltonian of the system and the detector equals to
118: \begin{equation}
119: \hat{H}=\hat{H}_S+\hat{H}_D+\hat{H}_I
120: \end{equation}
121: where $\hat{H}_D$ is the Hamiltonian of the detector and $\hat{H}_I$ represents
122: the interaction between the detector and the system. We choose the operator
123: $\hat{H}_I$ in the form
124: \begin{equation}
125: \hat{H}_I=\lambda\hat{q}\hat{H}_0
126: \end{equation}
127: where $\hat{q}$ is the operator acting in the Hilbert space of the detector and
128: the parameter $\lambda$ describes the strength of the interaction. This
129: system---detector interaction is that considered by von Neumann \cite{vNeum}
130: and in Refs. \cite{joos,caves,milb,gagen,rus}. In order to obtain a sensible
131: measurement, the parameter $\lambda$ must be large. We require a continuous
132: spectrum of operator $\hat{q}$. For simplicity, we can consider the quantity
133: $q$ as the coordinate of the detector.
134:
135: The measurement begins at time moment $t_0$. At the beginning of the
136: interaction with the detector, the detector is in the pure state $\left|\Phi
137: \right\rangle$. The full density matrix of the system and detector is
138: $\hat{\rho}(t_0)=\hat{\rho}_S(t_0)\otimes\left|\Phi\right\rangle\left\langle
139: \Phi\right|$ where $\hat{\rho}_S(t_0)$ is the density matrix of the system. The
140: duration of the measurement is $\tau$. After the measurement the density matrix
141: of the system is $\hat{\rho}_S(\tau+t_0)=Tr_D\left\{\hat{U}(\tau+t_0)\left(
142: \hat{\rho}_S(t_0)\otimes\left|\Phi\right\rangle\left\langle\Phi\right|\right)
143: \hat{U}^{\dag}(\tau+t_0)\right\}$ and the density matrix of the detector is
144: $\hat{\rho}_D(\tau+t_0)=Tr_S\left\{\hat{U}(\tau+t_0)\left(\hat{\rho}_S(t_0)\otimes
145: \left|\Phi\right\rangle\left\langle\Phi\right|\right)\hat{U}^{\dag}(\tau+t_0)\right\}$
146: where $\hat{U}(t)$ is the evolution operator of the system and detector,
147: obeying the equation
148: \begin{equation}
149: i\hbar\frac{\partial}{\partial t}\hat{U}(t)=\hat{H}(t)\hat{U}(t)
150: \end{equation}
151: with the initial condition $\hat{U}(t_0)=1$.
152:
153: Since the initial density matrix is chosen in a factorizable form, the density
154: matrix of the system after the interaction depends linearly on the density
155: matrix of the system before the interaction. We can represent this fact by the
156: equality
157: \begin{equation}
158: \hat{\rho}_S(\tau+t_0)=S(\tau,t_0)\hat{\rho}_S(t_0)
159: \label{eq:supop}
160: \end{equation}
161: where $S(\tau,t_0)$ is the superoperator acting on the density matrices of the
162: system. If the vectors $\left|n\right\rangle$ form the complete basis in the
163: Hilbert space of the system we can rewrite Eq.\ (\ref{eq:supop}) in the form
164: \begin{equation}
165: \rho_S(\tau+t_0)_{pr}=S(\tau,t_0)_{pr}^{nm}\rho_S(t_0)_{nm}
166: \label{eq:rho1}
167: \end{equation}
168: where the sum over the repeating indices is supposed. The matrix elements of
169: the superoperator are
170: \begin{equation}
171: S(\tau,t_0)_{pr}^{nm}=Tr_D\left\{\left\langle p\right|\hat{U}
172: (\tau+t_0)\left(\left|n\right\rangle\left\langle m\right|\otimes
173: \left|\Phi\right\rangle\left\langle\Phi\right|\right)\hat{U}^{\dag}(\tau+t_0)
174: \left|r\right\rangle\right\}.
175: \end{equation}
176:
177: Due to the finite duration of the measurement it is impossible to realize the
178: infinitely frequent measurements. The highest frequency of the measurements is
179: achieved when the measurements are performed one after another without the
180: period of the measurement-free evolution between two successive measurements.
181: Therefore, we model a continuous measurement by the subsequent measurements of
182: the finite duration and finite accuracy. After $N$ measurements the density
183: matrix of the system is
184: \begin{equation}
185: \hat{\rho}_S(N\tau)=S(\tau,(N-1)\tau)\ldots S(\tau,\tau)
186: S(\tau,0)\hat{\rho}_S(0).
187: \end{equation}
188:
189: Further, for simplicity we will neglect the Hamiltonian of the detector. After
190: this assumption the evolution operator is equal to
191: $\hat{U}(t,1+\lambda\hat{q})$ where the operator $\hat{U}(t,\xi)$ obeys the
192: equation
193: \begin{equation}
194: i\hbar\frac{\partial}{\partial t}\hat{U}(t,\xi)=\left(\xi
195: \hat{H}_0+\hat{H}_1+\hat{V}(t+t_0)\right)\hat{U}(t,\xi)
196: \label{eq:evol1}
197: \end{equation}
198: with the initial condition $\hat{U}(t_0,\xi)=1$. Then the superoperator
199: $S(\tau,t_0)$ is
200: \begin{equation}
201: S(\tau,t_0)_{pr}^{nm}=\int dq\left|\langle q\left|\Phi\right\rangle
202: \right|^2\left\langle p\right|\hat{U}(\tau+t_0,1+\lambda q)\left|n
203: \right\rangle\left\langle m\right|\hat{U}^{\dag}(\tau+t_0,1+\lambda q)
204: \left|r\right\rangle.
205: \label{eq:supmatr}
206: \end{equation}
207:
208: \section{Measurement of the unperturbed system}
209: \label{sec:id}
210:
211: In order to estimate the necessary duration of the single measurement it is
212: convenient to consider the case when the operator $\hat{V}=0$. In such a case
213: the description of the evolution is simpler. The measurement of this kind
214: occurs also when the influence of the perturbation operator $\hat{V}$ is small
215: in comparison with the interaction between the system and the detector and,
216: therefore, the operator $\hat{V}$ can be neglected.
217:
218: We can choose the basis $\left|n\alpha\right\rangle$ common for the operators
219: $\hat{H}_0$ and $\hat{H}_1$,
220: \begin{eqnarray}
221: \hat{H}_0\left|n\alpha\right\rangle&=&E_n\left|n\alpha\right\rangle\, \\
222: \hat{H}_1\left|n\alpha\right\rangle&=&E_1(n,\alpha)\left|n\alpha\right\rangle
223: \end{eqnarray}
224: where $n$ numbers the eigenvalues of the Hamiltonian $\hat{H}_0$ and $\alpha$
225: represents the remaining quantum numbers. Since the Hamiltonian of the system
226: does not depend on $t$ we will omit the parameter $t_0$ in this section. From
227: Eq.\ (\ref{eq:supmatr}) we obtain the superoperator $S(\tau)$ in the basis
228: $\left|n\alpha\right\rangle$
229: \begin{eqnarray}
230: S(\tau)_{p\alpha_3,r\alpha_4}^{n\alpha_1,m\alpha_2}&=&\delta_{np}
231: \delta_{mr}\delta(\alpha_1,\alpha_3)\delta(\alpha_2,\alpha_4)
232: \exp(i\omega_{m\alpha_2,n\alpha_1}\tau)\nonumber \\
233: &&\times\int dq\left|\langle q\left|\Phi\right\rangle\right|^2\exp(
234: i\lambda\omega_{mn}\tau q)
235: \label{eq:s1}
236: \end{eqnarray}
237: where
238: \begin{eqnarray}
239: \omega_{mn}&=&\frac{1}{\hbar}(E_m-E_n)\,, \\
240: \omega_{m\alpha_2,n\alpha_1}&=&\omega_{mn}+\frac{E_1(m,\alpha_2)-E_1(
241: n,\alpha_1)}{\hbar}
242: \end{eqnarray}
243: and $\delta({\cdot},{\cdot})$ represent the Kronecker's delta in a discrete
244: case and the Dirac's delta in a continuous case. Eq.\ (\ref{eq:s1}) can be
245: rewritten using the correlation function
246: \begin{equation}
247: F(\nu)=\left\langle\Phi\right|\exp(i\nu\hat{q})\left|\Phi\right\rangle.
248: \label{F}
249: \end{equation}
250: We can express this function as $F(\nu)=\int dq\left|\langle q\left|\Phi
251: \right\rangle\right|^2\exp(i\nu q)=\int dp\left\langle\Phi\right|p\rangle
252: \langle p-\frac{\nu}{\hbar}\left|\Phi\right\rangle$. Since vector $\left|
253: \Phi\right\rangle$ is normalized, the function $F(\nu)$ tends to zero when
254: $\left|\nu\right|$ increases. There exists a constant $C$ such that the
255: correlation function $\left|F(\nu)\right|$ is small if the variable
256: $\left|\nu\right|>C$.
257:
258: Then the equation for the superoperator $S(\tau)$ is
259: \begin{equation}
260: S(\tau)_{p\alpha_3,r\alpha_4}^{n\alpha_1,m\alpha_2}=\delta_{np}
261: \delta_{mr}\delta(\alpha_1,\alpha_3)\delta(\alpha_2,\alpha_4)
262: \exp(i\omega_{m\alpha_2,n\alpha_1}\tau)F(\lambda \tau\omega_{mn}).
263: \label{eq:s2}
264: \end{equation}
265: Using Eqs.\ (\ref{eq:rho1}) and\ (\ref{eq:s2}) we find that after the
266: measurement the non-diagonal elements of the density matrix of the system
267: become small, since $F(\lambda \tau\omega_{mn})$ is small for $n\neq m$ when
268: $\lambda \tau$ is large.
269:
270: The density matrix of the detector is
271: \begin{equation}
272: \left\langle q\right|\hat{\rho}_D(\tau)\left|q_1\right\rangle=\langle
273: q\left|\Phi\right\rangle\left\langle\Phi\right|q_1\rangle Tr\left\{
274: \hat{U}(\tau,1+\lambda q)\hat{\rho}_S(0)\hat{U}^{\dag}(\tau,1+\lambda
275: q_{1})\right\}.\label{eq:det1}
276: \end{equation}
277: From Eqs.\ (\ref{eq:evol1}) and\ (\ref{eq:det1}) we obtain
278: \begin{equation}
279: \left\langle q\right|\hat{\rho}_D(\tau)\left|q_1\right\rangle=\langle
280: q\left|\Phi \right\rangle\left\langle\Phi\right|q_1\rangle\sum_n\exp(i\lambda
281: \tau\omega_n(q_1-q))\sum_{\alpha}\left\langle n,\alpha\right|
282: \hat{\rho}_S(0)\left|n,\alpha\right\rangle
283: \end{equation}
284: where
285: \begin{equation}
286: \omega_n=\frac{1}{\hbar}E_n.
287: \end{equation}
288: The probability that the system is in the energy level $n$ may be expressed as
289: \begin{equation}
290: P(n)=\sum_{\alpha}\left\langle n,\alpha\right|\hat{\rho}_S(0)\left|n,\alpha
291: \right\rangle.
292: \end{equation}
293: Introducing the state vectors of the detector
294: \begin{equation}
295: \left|\Phi_E\right\rangle=\exp\left(-\frac{i}{\hbar}\lambda \tau
296: E\hat{q}\right) \left|\Phi\right\rangle
297: \end{equation}
298: we can express the density operator of the detector as
299: \begin{equation}
300: \hat{\rho}_D(\tau)=\sum_n\left|\Phi_{E_n}\right\rangle\left\langle\Phi_{E_n}\right|
301: P(n).
302: \end{equation}
303:
304: The measurement is complete when the states $\left|\Phi_E\right\rangle$ are
305: almost orthogonal. The different energies can be separated only when the
306: overlap between the corresponding states $\left|\Phi_E\right\rangle$ is almost
307: zero. The scalar product of the states $\left|\Phi_E\right\rangle$ with
308: different energies $E_1$ and $E_2$ is
309: \begin{equation}
310: \langle\Phi_{E_1}\left|\Phi_{E_2}\right\rangle=F(\lambda \tau\omega_{12}).
311: \end{equation}
312: The correlation function $\left|F(\nu)\right|$ is small when
313: $\left|\nu\right|>C$. Therefore, we have the estimation for the error of the
314: energy measurement $\Delta E$ as
315: \begin{equation}
316: \lambda \tau\Delta E\gtrsim\hbar C
317: \end{equation}
318: and we obtain the expression for the necessary duration of the measurement
319: \begin{equation}
320: \tau\gtrsim\frac{\hbar}{\Lambda\Delta E} \label{eq:tt}
321: \end{equation}
322: where
323: \begin{equation}
324: \Lambda=\frac{\lambda}{C}\,.
325: \end{equation}
326: Since in our model the measurements are performed immediately one after the
327: other, from Eq.~(\ref{eq:tt}) it follows that the rate of measurements is
328: proportional to the strength of the interaction $\lambda$ between the system
329: and the measuring device.
330:
331: \section{Measurement of the perturbed system}
332: \label{sec:meas}
333:
334: The operator $\hat{V}(t)$ represents the perturbation of the unperturbed
335: Hamiltonian $\hat{H}_0+\hat{H}_1$. We will take into account the influence of
336: the operator $\hat{V}$ by the perturbation method, assuming that the strength
337: of the interaction $\lambda$ between the system and detector is large.
338:
339: The operator $\hat{V}(t)$ in the interaction picture is
340: \begin{equation}
341: \tilde{V}(t,t_0,\xi)=\exp\left(\frac{i}{\hbar}(\xi\hat{H}_0+\hat{H}_1)t\right)
342: \hat{V}(t+t_0)\exp\left(-\frac{i}{\hbar}(\xi\hat{H}_0+\hat{H}_1)t\right).
343: \end{equation}
344: In the second order approximation the evolution operator equals to
345: \begin{eqnarray}
346: \hat{U}(\tau,t_0,\xi)&\approx&\exp\left(-\frac{i}{\hbar}(\xi\hat{H}_0+\hat{H}_1)\tau
347: \right)\left\{1+\frac{1}{i\hbar}\int_0^\tau dt\tilde{V}
348: (t,t_0,\xi)\right.\nonumber \\
349: &&-\left.\frac{1}{\hbar^2}\int_0^\tau dt_1\int_0^t
350: dt_2\tilde{V}(t_1,t_0,\xi)\tilde{V}(t_2,t_0\xi)\right\}.
351: \label{eq:evol2}
352: \end{eqnarray}
353: Using Eqs.\ (\ref{eq:supmatr}) and\ (\ref{eq:evol2}) we can obtain the
354: superoperator $S$ in the second order approximation, too. The expression for
355: the matrix elements of the superoperator $S$ is given in the appendix (Eqs.\
356: (\ref{eq:ap1}),\ (\ref{eq:ap2}) and\ (\ref{eq:ap3})).
357:
358: The probability of the jump from the level $|i\alpha\rangle$ to the level
359: $|f\alpha_1\rangle$ during the measurement is $W(i\alpha\rightarrow f\alpha_1)
360: =S(\tau,t_0)_{f\alpha_1,f\alpha_1}^{i\alpha,i\alpha}$. Using Eqs.\
361: (\ref{eq:ap1}),\ (\ref{eq:ap2}) and\ (\ref{eq:ap3}) we obtain
362: \begin{eqnarray}
363: W(i\alpha\rightarrow f\alpha_1)&=&\frac{1}{\hbar^2}\int_0^\tau dt_1\int_0^\tau
364: dt_2 F(\lambda\omega_{if}(t_2-t_1))V(t_1+t_0)_{f\alpha_1,i\alpha}
365: V(t_2+t_0)_{i\alpha,f\alpha_1}\nonumber \\
366: &&\times\exp\left(i\omega_{i\alpha,f\alpha_1}(t_2-t_1)\right).
367: \label{eq:W1}
368: \end{eqnarray}
369:
370: The expression for the jump probability can be further simplified if the
371: operator $\hat{V}$ does not depend on $t$. We introduce the function
372: \begin{equation}
373: \Phi(t)_{f\alpha_1,i\alpha}=\left|V_{f\alpha_1,i\alpha}\right|^2
374: \exp\left(\frac{i}{\hbar}(E_1(f,\alpha_1)-E_1(i,\alpha))t\right).
375: \end{equation}
376: Changing variables we can rewrite the jump probability as
377: \begin{equation}
378: W(i\alpha\rightarrow f\alpha_1)=\frac{2}{\hbar^2}\text{Re}\int_0^\tau dt
379: F(\lambda\omega_{fi}t)\exp(i\omega_{fi}t)(\tau-t) \Phi(t)_{f\alpha_1,i\alpha}.
380: \label{eq:W}
381: \end{equation}
382: Introducing the Fourier transformation of $\Phi(t)_{f\alpha_1,i\alpha}$
383: \begin{equation}
384: G(\omega)_{f\alpha_1,i\alpha}=\frac{1}{2\pi}\int_{-\infty}^{\infty}dt
385: \Phi(t)_{f\alpha_1,i\alpha}\exp(-i\omega t)
386: \end{equation}
387: and using Eq.\ (\ref{eq:W}) we obtain the equality
388: \begin{equation}
389: W(i\alpha\rightarrow f\alpha_1)=\frac{2\pi\tau}{\hbar^2}\int_{-\infty}^{\infty}
390: d\omega G(\omega)_{f\alpha_1,i\alpha}P(\omega)_{if}
391: \label{eq:resW}
392: \end{equation}
393: where
394: \begin{equation}
395: P(\omega)_{if}=\frac{1}{\pi}\text{Re}\int_0^\tau dt
396: F(\lambda\omega_{if}t)\exp(i(\omega-\omega_{if})t)
397: \left(1-\frac{t}{\tau}\right). \label{eq:pp}
398: \end{equation}
399: From Eq.\ (\ref{eq:pp}), using the equality $F(0)=1$, we obtain
400: \begin{equation}
401: \int d\omega P(\omega)_{if}=1.
402: \label{eq:prop3}
403: \end{equation}
404:
405: The quantity $G$ equals to
406: \begin{equation}
407: G(\omega)_{f\alpha_1,i\alpha}=\hbar\left|V_{f\alpha_1,i\alpha}\right|^2
408: \delta(E_1(f,\alpha_1)-E_1(i,\alpha)-\hbar\omega).
409: \label{eq:G}
410: \end{equation}
411: We see that the quantity $G(\omega)$ characterizes the perturbation.
412:
413: \section{The discrete spectrum}
414: \label{sec:discr}
415:
416: Let us consider the measurement effect on the system with the discrete
417: spectrum. The Hamiltonian $\hat{H}_0$ of the system has a discrete spectrum,
418: the operator $\hat{H}_1=0$, and the operator $\hat{V}(t)$ represents a
419: perturbation resulting in the quantum jumps between the discrete states of the
420: system $\hat{H}_0$.
421:
422: For the separation of the energy levels, the error in the measurement should be
423: smaller than the distance between the nearest energy levels of the system. It
424: follows from this requirement and Eq.\ (\ref{eq:tt}) that the measurement time
425: $\tau\gtrsim\frac{1}{\Lambda\omega_{\min}}$, where $\omega_{\min}$ is the
426: smallest of the transition frequencies $|\omega_{if}|$.
427:
428: When $\lambda$ is large then $\left|F(\lambda x)\right|$ is not very small only
429: in the region $\left|x\right|<\Lambda^{-1}$. We can estimate the probability of
430: the jump to the other energy level during the measurement, replacing $F(\nu)$
431: by $2C\delta(\nu)$in Eq.\ (\ref{eq:W1}). Then from Eq.\ (\ref{eq:W1}) we obtain
432: \begin{equation}
433: W(i\alpha\rightarrow f\alpha_1)\approx
434: \frac{2}{\hbar^2\Lambda\left|\omega_{if}\right|}\int_0^\tau dt \left|
435: V(t+t_0)_{i\alpha_1,f\alpha}\right|^2. \label{eq:jumpprob}
436: \end{equation}
437: We see that the probability of the jump is proportional to $\Lambda^{-1}$.
438: Consequently, for large $\Lambda$, i.e. for the strong interaction with the
439: detector, the jump probability is small. This fact represents the quantum Zeno
440: effect. However, due to the finiteness of the interaction strength the jump
441: probability is not zero. After sufficiently large number of measurements the
442: jump occurs. We can estimate the number of measurements $N$ after which the
443: system jumps into other energy levels from the equality
444: $\frac{2\tau}{\hbar^2\Lambda\left|\omega_{\min}\right|}\left|V_{\max}
445: \right|^2N\sim 1$ where $|V_{\max}|$ is the largest matrix element of the
446: perturbation operator $V$. This estimation allows us to introduce the
447: characteristic time, during which the evolution of the system is inhibited
448: \begin{equation}
449: t_{\text{inh}}\equiv \tau
450: N=\Lambda\frac{\hbar^2\left|\omega_{\min}\right|}{2\left| V_{\max}\right|^2}\,.
451: \label{eq:inht}
452: \end{equation}
453: We call this duration the inhibition time (it is natural to call this duration
454: the Zeno time, but this term has already different meaning).
455:
456: The full probability of the jump from level $|i\alpha\rangle$ to other levels
457: is $W(i\alpha)=\sum_{f,\alpha_1}W(i\alpha\rightarrow f\alpha_1)$. From Eq.\
458: (\ref{eq:jumpprob}) we obtain
459: \begin{equation}
460: W(i\alpha)=\frac{2}{\hbar^2\Lambda}\sum_{f,\alpha_1}
461: \frac{1}{\left|\omega_{if}\right|}\int_0^\tau dt
462: \left|V\left(t+t_0\right)_{f\alpha_1,i\alpha}\right|^2.
463: \end{equation}
464: If the matrix elements of the perturbation $V$ between different levels are of
465: the same size then the jump probability increases linearly with the number of
466: the energy levels. This behavior has been observed in Ref. \cite{Gagen2}.
467:
468: Due to the unitarity of the operator $\hat{U}(t,\xi)$ it follows from Eq.\
469: (\ref{eq:supmatr}) that the superoperator $S(\tau,t_0)$ obeys the equalities
470: \begin{mathletters}
471: \begin{eqnarray}
472: \sum_{p,\alpha}S(\tau,t_0)_{p\alpha,p\alpha}^{n\alpha_1,m\alpha_2}
473: &=&\delta_{nm}\delta_{\alpha_1,\alpha_2}, \\
474: \sum_{n,\alpha}S(\tau,t_0)_{p\alpha_1,r\alpha_2}^{n\alpha,n\alpha}
475: &=&\delta_{pr}\delta_{\alpha_1,\alpha_2}. \label{eq:prop1}
476: \end{eqnarray}
477: \end{mathletters}
478: If the system has a finite number of energy levels, the density matrix of the
479: system is diagonal and all states are equally occupied (i.e.,
480: $\rho(t_0)_{n\alpha_1,m\alpha_2}=\frac{1}{K}
481: \delta_{nm}\delta_{\alpha_1,\alpha_2}$ where $K$ is the number of the energy
482: levels) then from Eq.\ (\ref{eq:prop1}) it follows that
483: $S(\tau,t_0)\rho(t_0)=\rho(t_0)$. Such a density matrix is the stable point of
484: the map $\rho\rightarrow S\rho$. Therefore, we can expect that after a large
485: number of measurements the density matrix of the system tends to this density
486: matrix.
487:
488: When $\Lambda$ is large and the duration of the measurement is small, we can
489: neglect the non-diagonal elements in the density matrix of the system, since
490: they always are of order $\Lambda^{-1}$. Replacing $F(\nu)$ by $2C\delta(\nu)$
491: in Eqs.\ (\ref{eq:ap1}),\ (\ref{eq:ap2}) and\ (\ref{eq:ap3}) and neglecting the
492: elements of the superoperator $S$ that cause the arising of the non-diagonal
493: elements of the density matrix, we can write the equation for the
494: superoperator $S$ as
495: \begin{equation}
496: S(\tau,t_0)_{p\alpha_3,r\alpha_4}^{n\alpha_1,m\alpha_2}\approx
497: \delta_{pn}\delta(\alpha_3,\alpha_1)\delta_{rm}
498: \delta(\alpha_4,\alpha_2)\delta_{pr}+\frac{1}{\Lambda}
499: A(\tau,t_0)_{p,\alpha_3,\alpha_4}^{n,\alpha_1,\alpha_2}\delta_{pr}\delta_{nm}
500: \end{equation}
501: where
502: \begin{eqnarray}
503: A(\tau,t_0)_{p,\alpha_3,\alpha_4}^{n,\alpha_1,\alpha_2}
504: &=&\frac{2}{\hbar^2\left|\omega_{np}\right|}\int_0^\tau dt
505: V(t+t_0)_{p\alpha_3,n\alpha_1}V(t+t_0)_{n\alpha_2,p\alpha_4}\nonumber \\
506: &&-\delta_{pn}\delta(\alpha_4,\alpha_2)\sum_{s,\alpha}
507: \frac{1}{\hbar^2\left|\omega_{sn}\right|}\int_0^\tau dt
508: V(t+t_0)_{n\alpha_3,s\alpha}V(t+t_0)_{s\alpha,n\alpha_1}
509: \nonumber \\
510: &&-\delta_{pn}\delta(\alpha_3,\alpha_1)\sum_{s,\alpha}
511: \frac{1}{\hbar^2\left|\omega_{ns}\right|}\int_0^\tau dt V(t+t_0
512: )_{s\alpha,n\alpha_4}V(t+t_0)_{n\alpha_2,s\alpha}
513: \end{eqnarray}
514: Then for the diagonal elements of the density matrix we have
515: $\rho(\tau+t_0)\approx\rho(t_0)+\frac{1}{\Lambda}A(\tau,t_0)\rho(t_0)$, or
516: \begin{equation}
517: \frac{d}{dt}\hat{\rho}(t)\approx\frac{1}{\Lambda\tau}A(\tau,t) \hat{\rho}(t).
518: \label{eq:deriv}
519: \end{equation}
520: If the perturbation $V$ does not depend on $t$ then it follows from
521: Eq.~(\ref{eq:deriv}) that the diagonal elements of the density matrix evolve
522: exponentially.
523:
524: \subsection{Example}
525:
526: As an example we will consider the evolution of the measured two-level system.
527: The system is forced by the perturbation $V$ which induces the jumps from one
528: state to another. The Hamiltonian of this system is
529: \begin{equation}
530: \hat{H}=\hat{H}_0+\hat{V} \label{eq:ham1}
531: \end{equation}
532: where
533: \begin{eqnarray}
534: \hat{H}_0&=&\frac{\hbar\omega}{2}\hat{\sigma}_3, \label{eq:ham2}\\
535: \hat{V}&=&v\hat{\sigma}_{+}+v^{*}\hat{\sigma}_{-}. \label{eq:ham3}
536: \end{eqnarray}
537: Here $\sigma_1,\sigma_2,\sigma_3$ are Pauli matrices and $\sigma_{\pm}
538: =\frac{1}{2}(\sigma_1\pm i\sigma_2)$. The Hamiltonian $\hat{H}_0$ has two
539: eigenfunctions $\left|0\right\rangle$ and $\left|1\right\rangle$ with the
540: eigenvalues $-\hbar\frac{\omega}{2}$ and $\hbar\frac{\omega}{2}$ respectively.
541: The evolution operator of the unmeasured system is
542: \begin{equation}
543: \hat{U}(t)=\cos\left(\frac{\Omega}{2}t\right)-\frac{2i}{\hbar\Omega}\hat{H}
544: \sin\left(\frac{\Omega}{2}t\right)
545: \label{eq:evol4}
546: \end{equation}
547: where
548: \begin{equation}
549: \Omega=\sqrt{\omega^2+4\frac{\left|v\right|^2}{\hbar^2}}.
550: \end{equation}
551: If the initial density matrix is $\rho(0)=|1\rangle\langle 1|$ then the
552: evolution of the diagonal elements of the unmeasured system's density matrix is
553: given by the equations
554: \begin{mathletters}
555: \begin{eqnarray}
556: \rho_{11}(t)&=&\cos^2\left(\frac{\Omega}{2}
557: t\right)+\left(\frac{\omega}{\Omega}\right)^2\sin^2\left(\frac{\Omega}{2}t\right)
558: \label{eq:free}\\
559: \rho_{00}(t)&=&\left(1-\left(\frac{\omega}{\Omega}\right)^2\right)\sin^2\left(
560: \frac{\Omega}{2}t\right).
561: \end{eqnarray}
562: \end{mathletters}
563:
564: Let us consider now the dynamics of the measured system. The equations for the
565: diagonal elements of the density matrix (Eq.\ (\ref{eq:deriv}) ) for the system
566: under consideration are
567: \begin{mathletters}
568: \label{eq:eqs1}
569: \begin{eqnarray}
570: \frac{d}{dt}\rho_{11} &\approx &-\frac{1}{t_{\text{inh}}}\left(\rho_{11}
571: -\rho_{00}\right), \\
572: \frac{d}{dt}\rho_{00} &\approx &-\frac{1}{t_{\text{inh}}}\left(\rho_{00}
573: -\rho_{11}\right).
574: \end{eqnarray}
575: \end{mathletters}
576: where the inhibition time, according to Eq.\ (\ref{eq:inht}), is
577: \begin{equation}
578: t_{\text{inh}}=\frac{\Lambda}{2\omega}\left|\frac{\hbar\omega}{v}\right|^2.
579: \end{equation}
580: The solution of Eqs.\ (\ref{eq:eqs1}) with the initial condition
581: $\rho(0)=\left|1\right\rangle\left\langle 1\right|$ is
582: \begin{mathletters}
583: \begin{eqnarray}
584: \rho_{11}(t) &=&\frac{1}{2}\left(1+\exp\left(-\frac{2}{t_{
585: \text{inh}}}t\right)\right), \label{eq:aprox1}\\
586: \rho_{00}(t) &=&\frac{1}{2}\left(1-\exp\left(-\frac{2}{t_{
587: \text{inh}}}t\right)\right).
588: \end{eqnarray}
589: \end{mathletters}
590:
591: From Eq.~(\ref{eq:prop1}) it follows that if the density matrix of the system
592: is
593: \begin{equation}
594: \hat{\rho}_f=\frac{1}{2}\left(\left|0\right\rangle\left\langle 0\right|
595: +\left|1\right\rangle\left\langle 1\right|\right),
596: \end{equation}
597: then $S(\tau)\hat{\rho}_f=\hat{\rho}_f$. Hence, when the number of the
598: measurements tends to infinity, the density matrix of the system approaches
599: $\hat{\rho}_f$.
600:
601: \begin{figure}
602: \begin{center}
603: \epsfxsize=.6\hsize
604: \epsffile{fig1.eps}
605: \end{center}
606: \caption{The occupation of the initial level $1$ of the measured two-level
607: system calculated according to Eqs.\ (\protect\ref{eq:rho1}),\
608: (\protect\ref{eq:supmatr}),\ (\protect\ref{eq:evol4}) and\
609: (\protect\ref{eq:fcalc}). The used parameters are $\hbar=1$, $\sigma^2=1$,
610: $\omega=2$, $v=1$. The strength of the measurement $\lambda=50$ and the
611: duration of the measurement $\tau=0.1$. The exponential approximation
612: (\protect\ref{eq:aprox1}) is shown as a dashed line. For comparison the
613: occupation of the level $1$ of the unmeasured system is also shown (dotted
614: line).}
615: \label{fig1}
616: \end{figure}
617: \begin{figure}
618: \begin{center}
619: \epsfxsize=.6\hsize
620: \epsffile{fig2.eps}
621: \end{center}
622: \caption{The non-diagonal element of the density matrix of the measured
623: two-level system. Used parameters are the same as in Fig. \protect\ref{fig1}}
624: \label{fig2}
625: \end{figure}
626: \begin{figure}
627: \begin{center}
628: \epsfxsize=.6\hsize
629: \epsffile{fig3.eps}
630: \end{center}
631: \caption{The occupation of the initial level $1$ of the measured two-level
632: system for different strengths of the measurement: $\lambda=50$, $\tau=0.1$
633: (dashed line) and $\lambda=5$, $\tau=0.2$ (solid line). Other parameters are
634: the same as in Fig. \protect\ref{fig1}}
635: \label{fig3}
636: \end{figure}
637:
638: We have performed the numerical analysis of the dynamics of the measured
639: two-level system (\ref{eq:ham1})---(\ref{eq:ham3}) using Eqs.\
640: (\ref{eq:rho1}),\ (\ref{eq:supmatr}) and\ (\ref{eq:evol4}) with the Gaussian
641: correlation function (\ref{F})
642: \begin{equation}
643: F(\nu)=\exp\left(-\frac{\nu^2}{2\sigma^2}\right). \label{eq:fcalc}
644: \end{equation}
645: From the condition $\int_{-\infty}^{\infty}F(\nu)d\nu=2C$ we have
646: $C=\sigma\sqrt{\frac{\pi}{2}}$. The initial state of the system is
647: $\left|1\right\rangle$. The matrix elements of the density matrix
648: $\rho(t)_{11}$ and $\rho(t)_{10}$ are represented in Fig. \ref{fig1} and Fig.
649: \ref{fig2}, respectively. In Fig. \ref{fig1} the approximation
650: (\ref{eq:aprox1}) is also shown. This approach is close to the exact evolution.
651: The matrix element $\rho(t)_{11}$ for two different values of $\lambda$ is
652: shown in Fig. \ref{fig3}. We see that for larger $\lambda$ the evolution of the
653: system is slower.
654:
655: The influence of the repeated non-ideal measurements on the two level system
656: driven by the periodic perturbation has also been considered in Refs.
657: \cite{Gagen2,Peres,Jordan,Venugoplan}. Similar results have been found: the
658: occupation of the energy levels changes exponentially with time, approaching
659: the limit $\frac{1}{2}$.
660:
661: \section{The decaying system}
662: \label{sec:dec}
663:
664: We consider the system which consists of two parts. We can treat the first part
665: as an atom, and the second part as the field (reservoir). The energy spectrum
666: of the atom is discrete and the spectrum of the field is continuous. The
667: Hamiltonians of these parts are $\hat{H}_0$ and $\hat{H}_1$ respectively and
668: the eigenfunctions are $\left|n\right\rangle$ and $\left|\alpha\right\rangle$,
669: \begin{mathletters}
670: \begin{eqnarray}
671: \hat{H}_0\left|n\right\rangle&=&E_{n}\left|n\right\rangle, \\
672: \hat{H}_1\left|\alpha\right\rangle&=&E_{\alpha}\left|\alpha\right\rangle.
673: \end{eqnarray}
674: \end{mathletters}
675: There is the interaction between the atom and the field represented by the
676: operator $\hat{V}$. So, the Hamiltonian of the system is
677: \begin{equation}
678: \hat{H}_{S}=\hat{H}_{0}+\hat{H}_{1}+\hat{V}.
679: \end{equation}
680: The basis for the full system is $\left|n\alpha\right\rangle=\left|
681: n\right\rangle\otimes\left|\alpha\right\rangle$.
682:
683: When the measurement is not performed, such a system exhibits exponential
684: decay, valid for the intermediate times. The decay rate is given according to
685: the Fermi's Golden Rule
686: \begin{equation}
687: R(i\alpha_1\rightarrow f\alpha_2)=\frac{2\pi}{\hbar}
688: \left|V_{f\alpha_2,i\alpha_1}\right|^2\rho\left(\hbar\omega_{if}\right)
689: \label{eq:goldrule}
690: \end{equation}
691: where
692: \begin{equation}
693: \frac{1}{\hbar}(E_{\alpha_2}-E_{\alpha_1})=\omega_{if}
694: \end{equation}
695: and $\rho\left(E\right)$ is the density of the reservoir's states.
696:
697: When the energy level of the atom is measured, we can use the perturbation
698: theory, as it is in the discrete case.
699:
700: The initial state of the field is a vacuum state $\left|0\right\rangle$
701: with energy $E_0=0$. Then the density matrix of the atom is $\hat{\rho}_0
702: (\tau)=Tr_1\left\{\hat{\rho}(\tau)\right\}=Tr_1
703: \left\{S(\tau)\hat{\rho}(0)\right\}$ or
704: $\hat{\rho}_0(\tau)=S_{ef}(\tau)\hat{\rho}_0(0)$, where $S_{ef}$ is an
705: effective superoperator
706: \begin{equation}
707: S_{ef}(\tau)_{pr}^{nm}=\sum_{\alpha}S(\tau)_{p\alpha,r\alpha}^{n0,m0}.
708: \label{eq:sef}
709: \end{equation}
710: When the states of the atom are weakly coupled to a broad band of states
711: (continuum), the transitions back to the excited state of the atom can be
712: neglected (i.e., we neglect the influence of emitted photons on the atom).
713: Therefore, we can use the superoperator $S_{ef}$ for determination of the
714: evolution of the atom.
715:
716: Since the states in the reservoir are very dense, one can replace the sum over
717: $\alpha $ by an integral over $E_{\alpha}$
718: \[
719: \sum_{\alpha}\ldots=\int dE_{\alpha}\rho(E_{\alpha})\ldots
720: \]
721: where $\rho(E_{\alpha})$ is the density of the states in the reservoir.
722:
723: \subsection{The spectrum}
724:
725: The density matrix of the field is $\hat{\rho}_1(\tau)=Tr_0\left\{
726: \hat{\rho}(\tau)\right\}=Tr_0\left\{S(\tau)\hat{\rho}(0)\right\}$. The diagonal
727: elements of the field's density matrix give the spectrum. If the initial state
728: of the atom is $\left|i\right\rangle$ then the distribution of the field's
729: energy is $W\left(E_{\alpha}\right)=\rho_1(\tau)_{\alpha\alpha}=
730: \sum_{f}S(\tau)_{f\alpha,f\alpha}^{i0,i0}$ . From Eqs.\ (\ref{eq:ap1}),\
731: (\ref{eq:ap2}) and\ (\ref{eq:ap3}) we obtain
732: \begin{equation}
733: W(E_{\alpha})=\sum_{f}\frac{2\pi}{\hbar^2}\left|V_{f\alpha,i0}\right|^2\tau
734: P\left(\frac{E_{\alpha}}{\hbar}\right)_{if}
735: \label{eq:spectr}
736: \end{equation}
737: where $P(\omega)_{if}$ is given by the equation (\ref{eq:pp}).
738: %\[
739: %P(\omega)_{if}=\frac{1}{\pi}\text{Re}\int_0^\tau dt
740: %F(\lambda\omega_{if}t)\exp(i(\omega-\omega_{if})t)
741: %\left(1-\frac{t}{\tau}\right)
742: %\]
743: From Eq.\ (\ref{eq:spectr}) we see that $P(\omega)$ is the measurement-modified
744: shape of the spectral line.
745:
746: The integral in Eq.\ (\ref{eq:pp}) is small when the exponent oscillates more
747: rapidly than the function $F$. This condition is fulfilled when
748: $\frac{E}{\hbar}-\omega_{if}\gtrsim\frac{\lambda\omega_{if}}{C}$. Consequently,
749: the width of the spectral line is
750: \begin{equation}
751: \Delta E_{if}=\Lambda\hbar\omega_{if}.
752: \end{equation}
753: The width of the spectral line is proportional to the strength of the
754: measurement (this equation is obtained using the assumption that the strength
755: of the interaction with the measuring device $\lambda$ is large and, therefore,
756: the natural width of the spectral line can be neglected). The broadening of the
757: spectrum of the measured system is also reported in Ref. \cite{Elattari} for
758: the case of an electron tunneling out of a quantum dot.
759:
760: \subsection{The decay rate}
761:
762: The probability of the jump from the state $i$ to the state $f$ is
763: $W(i\rightarrow f;\tau)=S_{ef}(\tau)_{ff}^{ii}$. From Eqs.\ (\ref{eq:sef}) it
764: follows
765: \begin{equation}
766: W(i\rightarrow f;\tau)=\sum_{\alpha}W(i0,\rightarrow f\alpha,\tau)
767: \end{equation}
768: Using Eq.\ (\ref{eq:resW}) we obtain the equality
769: \begin{equation}
770: W\left(i\rightarrow f;\tau\right)=\frac{2\pi
771: \tau}{\hbar^{2}}\int_{-\infty}^{+\infty} d\omega
772: G\left(\omega\right)_{fi}P\left(\omega\right)_{if}.
773: \label{eq:decprob2}
774: \end{equation}
775: where
776: \begin{equation}
777: G(\omega)_{fi}=\int dE_{\alpha}\rho(E_{\alpha})G(\omega)_{f\alpha,i0}
778: \end{equation}
779: The expression for $G(\omega)$ according to Eq.\ (\ref{eq:G}) is
780: \begin{equation}
781: G(\omega)_{fi}=\hbar\rho(\hbar\omega)\left|V_{fE_{\alpha}=\hbar\omega
782: ,i0}\right|^2.
783: \end{equation}
784: The quantity $G\left(\omega\right)$ is the reservoir coupling spectrum.
785:
786: The measurement-modified decay rate is $R\left(i\rightarrow f\right)=
787: \frac{1}{\tau}W(i\rightarrow f;\tau)$. From Eq.\ (\ref{eq:decprob2}) we have
788: \begin{equation}
789: R(i\rightarrow f)=\frac{2\pi}{\hbar^2}\int_{-\infty}^{\infty}d\omega
790: G(\omega)_{fi}P(\omega)_{if}. \label{eq:result}
791: \end{equation}
792: The equation\ (\ref{eq:result}) represents a universal result: the decay rate
793: of the frequently measured decaying system is determined by the overlap of the
794: reservoir coupling spectrum and the measurement-modified level width. This
795: equation was derived by Kofman and Kurizki \cite{Kofman2}, assuming the ideal
796: instantaneous projections. We show that Eq.\ (\ref{eq:result}) is valid for the
797: more realistic model of the measurement, as well. An equation, similar to Eq.\
798: (\ref{eq:result}) has been obtained in Ref. \cite{Panov}, considering a
799: destruction of the final decay state.
800:
801: Depending on the reservoir spectrum $G(\omega)$ and the strength of the
802: measurement the inhibition or acceleration of the decay can be obtained. If the
803: interaction with the measuring device is weak and, consequently, the width of
804: the spectral line is much smaller than the width of the reservoir spectrum, the
805: decay rate equals the decay rate of the unmeasured
806: system, given by the Fermi's Golden Rule\ (\ref{eq:goldrule}). In the
807: intermediate region, when the width of the spectral line is rather small
808: compared with the distance between $\omega_{if}$ and the nearest maximum in
809: the reservoir spectrum, the decay rate grows with increase of $\Lambda$. This
810: results in the anti-Zeno effect.
811:
812: If the width of the spectral line is much greater compared both with the width
813: of the reservoir spectrum and the distance between $\omega_{if}$ and the
814: centrum of the reservoir spectrum, the decay rate decreases when $\Lambda$
815: increases. This results in the quantum Zeno effect. In such a case we can use
816: the approximation
817: \begin{equation}
818: G\left(\omega\right)_{fi}\approx\hbar B_{fi}\delta\left(\omega-\omega_R
819: \right).
820: \end{equation}
821: where $B_{fi}$ is defined by the equality $B_{fi}=\frac{1}{\hbar}\int
822: G\left(\omega\right)_{fi}d\omega$ and $\omega_R$ is the centrum of $G(\omega)$.
823: Then from Eq.\ (\ref{eq:result}) we obtain the decay rate $R\left(i\rightarrow
824: f\right)\approx\frac{2\pi}{\hbar}B_{fi} P\left(\omega_{if}\right)_{if}$. From
825: Eq.\ (\ref{eq:pp}), using the condition $\Lambda\tau
826: \left|\omega_{if}\right|\gg 1$ and the equality
827: $\int_{-\infty}^{\infty}F\left(\nu\right)d\nu=2C$ we obtain
828: %the measurement-modified shape of the spectral line
829: \begin{equation}
830: P\left(\omega_{if}\right)_{if}=\frac{1}{\pi\Lambda\omega_{if}}.
831: \end{equation}
832: Therefore, the decay rate is equal to
833: \begin{equation}
834: R\left(i\rightarrow f\right)\approx\frac{2 B_{fi}}{\Lambda\hbar\omega_{if}}.
835: \end{equation}
836: The obtained decay rate is insensitive to the spectral shape of the reservoir
837: and is inverse proportional to the measurement strength $\Lambda$.
838:
839: \section{Summary}
840: \label{sec:concl}
841:
842: In this work we investigate the quantum Zeno effect using the definite model of
843: the measurement. We take into account the finite duration and the finite
844: accuracy of the measurement. The general equation for the probability of the
845: jump during the measurement is derived (\ref{eq:resW}). The behavior of the
846: system under the repeated measurements depends on the strength of measurement
847: and on the properties of the system.
848:
849: When the the strength of the interaction with the measuring device is
850: sufficiently large, the frequent measurements of the system with discrete
851: spectrum slow down the evolution. However, the evolution cannot be fully
852: stopped. Under the repeated measurements the occupation of the energy levels
853: changes exponentially with time, approaching the limit of the equal occupation
854: of the levels. The jump probability is inversely proportional to the strength
855: of the interaction with the measuring device.
856:
857: In the case of a continuous spectrum the measurements can cause inhibition or
858: acceleration of the evolution. Our model of the continuous measurement gives
859: the same result as the approach based on the projection postulate
860: \cite{Kofman2}. The decay rate is equal to the convolution of the reservoir
861: coupling spectrum with the measurement-modified shape of the spectral line. The
862: width of the spectral line is proportional to the strength of the interaction
863: with the measuring device. When this width is much greater than the width of
864: the reservoir, the quantum Zeno effect takes place. Under these conditions the
865: decay rate is inversely proportional to the strength of the interaction with
866: the measuring device. In a number of decaying systems, however, the reservoir
867: spectrum $G(\omega)$ grows with frequency almost up to the relativistic cut-off
868: and the strength of the interaction required for the appearance of the quantum
869: Zeno effect is so high that the initial system is significantly modified. When
870: the spectral line is not very broad, the decay rate may be increased by the
871: measurements more often than it may be decreased and the quantum anti-Zeno
872: effect can be obtained.
873:
874: \appendix
875: \section*{The superoperator}
876:
877: We obtain the superoperator $S$ in the second order approximation substituting
878: the approximate expression for the evolution operator (\ref{eq:evol2}) into
879: Eq.\ (\ref{eq:supmatr}). Thus we have
880: \begin{equation}
881: S(\tau,t_0)=S^{(0)}(\tau)+S^{(1)}(\tau,t_0)+S^{(2)}(\tau,t_0)\label{eq:ap1}
882: \end{equation}
883: where $S^{(0)}(\tau)$ is the superoperator of the unperturbed measurement given
884: by Eq.\ (\ref{eq:s2}), $S^{(1)}(\tau,t_0)$ is the first order correction,
885: \begin{eqnarray}
886: S^{(1)}(\tau,t_0)_{p\alpha_3,r\alpha_4}^{n\alpha_1,m\alpha_2}
887: &=&\frac{1}{i\hbar}\delta_{rm}\delta(\alpha_4,\alpha_2)\exp(i
888: \omega_{r\alpha_4,p\alpha_3}\tau)\int_0^\tau dt V(t+t_0)_{p\alpha_3,n\alpha_1} \nonumber \\
889: &&\times\exp(i\omega_{p\alpha_3,n\alpha_1}t)F(\lambda(\omega_{rp}\tau
890: +\omega_{pn}t)) \nonumber \\
891: &&-\frac{1}{i\hbar}\delta_{pn}\delta(\alpha_3,\alpha_1)\exp(i
892: \omega_{r\alpha_4,p\alpha_3}\tau)\int_0^\tau dt V(t+t_0)_{m\alpha_2,r\alpha_4} \nonumber \\
893: &&\times\exp(i\omega_{m\alpha_2,r\alpha_4}t)F(\lambda(\omega_{rp}\tau
894: +\omega_{mr}t)),
895: \label{eq:ap2}
896: \end{eqnarray}
897: and $S^{(2)}(\tau,t_0)$ is the second order correction,
898: \begin{eqnarray}
899: S^{(2)}(\tau,t_0)_{p\alpha_3,r\alpha_4}^{n\alpha_1,m\alpha_2}
900: &=&\frac{1}{\hbar^2}\exp(i\omega_{r\alpha_4,p\alpha_3}\tau)\int_0^\tau
901: dt_1\int_0^\tau dt_2V(t_1+t_0)_{p\alpha_3,n\alpha_1}
902: V(t_2+t_0)_{m\alpha_2,r\alpha_4} \nonumber \\
903: &&\times F(\lambda(\omega_{rp}\tau+\omega_{pn}t_1+\omega_{mr}t_2))
904: \exp(i\omega_{p\alpha_3,n\alpha_1}t_1
905: +i\omega_{m\alpha_2,r\alpha_4}t_2) \nonumber \\
906: &&-\frac{1}{\hbar^2}\delta_{rm}\delta(\alpha_4,\alpha_2)\exp(i
907: \omega_{r\alpha_4,p\alpha_3}\tau)\sum_{s,\alpha} \nonumber \\
908: &&\int_0^\tau dt_1\int_0^{t_1}dt_2 V(t_1+t_0)_{p\alpha_3,s\alpha}
909: V(t_2+t_0)_{s\alpha,n\alpha_1}\nonumber \\
910: &&\times F(\lambda(\omega_{rp}\tau+\omega_{ps}t_1+\omega_{sn}t_2
911: ))\exp(i\omega_{p\alpha_3,s\alpha}t_1
912: +i\omega_{s\alpha,n\alpha_1}t_2) \nonumber \\
913: &&-\frac{1}{\hbar^2}\delta_{pn}\delta(\alpha_3,\alpha_1)\exp(i
914: \omega_{r\alpha_4,p\alpha_3}\tau)\sum_{s,\alpha}\nonumber \\
915: &&\int_0^\tau dt_1\int_0^{t_1}dt_2 V(t_2+t_0)_{m\alpha_2,s\alpha}
916: V(t_1+t_0)_{s\alpha,r\alpha_4}\nonumber \\
917: &&\times F(\lambda(\omega_{rp}\tau+\omega_{sr}t_1+\omega_{ms}t_2
918: ))\exp(i\omega_{s\alpha,r\alpha_4}t_1+i\omega_{m\alpha_2,s\alpha}t_2),
919: \label{eq:ap3}
920: \end{eqnarray}
921: where
922: \begin{equation}
923: \omega_{n\alpha_1,m\alpha_2}=\omega_{nm}+\frac{E_1(n,\alpha_1)-
924: E_1(m,\alpha_2)}{\hbar}.
925: \end{equation}
926:
927: \begin{references}
928: \bibitem{vNeum} J. von Neumann, {\it Mathematisch Grundlagen der
929: Quanten-mechanik} (Springer, Berlin, 1932). English translation: {\it
930: Mathematical Foundations of Quantum Mechanics} (Princeton University Press,
931: Princeton, NJ, 1955).
932:
933: \bibitem{Khalfin} L. A. Khalfin, Zh. Eksp. Teor. Fiz. {\bf 33}, 1371 (1958)
934: [Sov. Phys. JETP {\bf 6}, 1503 (1958)]; L. Fonda, G. C. Ghirardi, and
935: A.~Rimini, Rep. Prog. Phys. {\bf 41}, 587 (1978).
936:
937: \bibitem{GG} S. R. Wilkinson, C. F. Bharucha, M. C. Fisher, K. W. Madison,
938: P. R. Morrow, Q. Niu, B. Sundaram, and M. G. Raizen, Nature {\bf 387}, 575
939: (1997).
940:
941: \bibitem{Mishra} B. Mishra and E. C. G. Sudarshan, J. Math. Phys. {\bf 18}, 756
942: (1977).
943:
944: \bibitem{Itano} W. M. Itano, D. Heinzen, J. J. Bollinger, and D. J. Wineland, Phys.
945: Rev. A {\bf 41}, 2295 (1990).
946:
947: \bibitem{Petrosky} T. Petrosky, S. Tasaki, and I. Prigogine, Phys. Lett. A {\bf
948: 151}, 109 (1990); T. Petrosky, S. Tasaki, and I. Prigogine, Physica A {\bf
949: 170}, 306 (1991).
950:
951: \bibitem{FS} V. Frerichs and A. Schenzle, Phys. Rev. A {\bf 44}, 1962 (1991).
952:
953: \bibitem{Namiki} S. Pascazio and M. Namiki, Phys. Rev. A {\bf 50}, 4582 (1994).
954:
955: \bibitem{Kofman1} A. G. Kofman and G. Kurizki, Phys. Rev. A {\bf 54}, R3750
956: (1996).
957:
958: \bibitem{Kaulakys} B. Kaulakys and V. Gontis, Phys. Rev. A {\bf 56}, 1131
959: (1997); V Gontis and B. Kaulakys, Lith. J. Phys. {\bf 38}, 118 (1998); e-print
960: quant-ph/9806015.
961:
962: \bibitem{Pascazio2} S. Pascazio and P. Facchi, e-print quant-ph/9904076.
963:
964: \bibitem{Elattari} B. Elattari and S. A. Gurvitz, e-print quant-ph/0001020.
965:
966: \bibitem{Facchi} P. Facchi, H. Nakazato, and S. Pascazio, e-print quant-ph/0006094.
967:
968: \bibitem{Kofman2} A. G. Kofman and G. Kurizki, Nature {\bf 405}, 546 (2000).
969:
970: \bibitem{Lewenstein} M. Lewenstein and K. Rz\c{a}\v{z}ewski, Phys. Rev. A {\bf
971: 61}, 022105 (2000).
972:
973: \bibitem{Fearn} H. Fearn and W. E. Lamb Jr, Phys. Rev. A {\bf 46}, 1199 (1992).
974:
975: \bibitem{prezhdo} O. V. Prezhdo, Phys. Rev. Lett {\bf 85}, 4413 (2000).
976:
977: \bibitem{joos} E. Joos, Phys. Rev. D {\bf 29}, 1626 (1984).
978:
979: \bibitem{caves} C. M. Caves and G. J. Milburn, Phys. Rev. A, {\bf 36}, 5548
980: (1987).
981:
982: \bibitem{milb} G. J. Milburn, J. Opt. Soc. Am. B {\bf 5}, 1317 (1988).
983:
984: \bibitem{gagen} M. J. Gagen, H. M. Wiseman, and G. J. Milburn, Phys. Rev. A
985: {\bf 48}, 132 (1993).
986:
987: \bibitem{rus} J. Ruseckas, Phys. Rev. A (to be published).
988:
989: \bibitem{Gagen2} M. J. Gagen and G. J. Milburn, Phys. Rev. A {\bf 45}, 5228
990: (1992).
991:
992: \bibitem{Peres} A. Peres and A. Ron, Phys. Rev. A {\bf 42}, 5720 (1990).
993:
994: \bibitem{Jordan} T. F. Jordan, E. C. G. Sudarshan, and P. Valanju, Phys. Rev. A
995: {\bf 44}, 3340 (1991).
996:
997: \bibitem{Venugoplan} A. Venugoplan and R. Ghosh, Phys. Lett. A {\bf 204}, 11
998: (1995).
999:
1000: \bibitem{Panov} A. D. Panov, Phys. Lett. A {\bf 260}, 441 (1999).
1001: \end{references}
1002:
1003: \end{document}
1004: