a

r
X
i
v
:
1
3
0
5
.
2
5
1
7
v
1


[
q
u
a
n
t
-
p
h
]


1
1

M
a
y

2
0
1
3
Dividing Line between Quantum and Classical Trajectories:
Bohmian Time Constant
Antonio B. Nassar
1
and Salvador Miret-Artes
2
1
Science Department, Harvard-Westlake School,
3700 Coldwater Canyon, Studio City, 91604, USA
1
Department of Sciences, University of California, Los Angeles, Extension Program
10995 Le Conte Avenue, Los Angeles, CA 90024, USA
2
Instituto de Fsica Fundamental, Consejo Superior de Investigaciones Cientcas,
Serrano 123, 28006 Madrid, Spain
Abstract
This work proposes an answer to a challenge posed by Bell on the lack of clarity in
regards to the line between the quantum and classical regimes in a measurement problem. To
this end, a generalized logarithmic nonlinear Schr odinger equation is proposed to describe the
time evolution of a quantum dissipative system under continuous measurement. Within the
Bohmian mechanics framework, a solution to this equation reveals a novel result: it displays
a time constant which should represent the dividing line between the quantum and classical
trajectories. It is shown that continuous measurements and damping not only disturb the
particle but compel the system to converge in time to a Newtonian regime. While the
width of the wave packet may reach a stationary regime, its quantum trajectories converge
exponentially in time to classical trajectories. In particular, it is shown that damping tends
to suppress further quantum eects on a time scale shorter than the relaxation time of
the system. If the initial wave packet width is taken to be equal to 2.8 10
15
m (the
approximate size of an electron), the Bohmian time constant is found to have an upper
limit, i. e.,
B max
= 10
26
s.
PACS: 03.65.Ta
E-mail: nassar@ucla.edu
2
As pointed out by Bell,[1] the lack of clarity in regards to where the transition between
the classical and quantum regimes is located is one aspect of the measurement problem. This
problem represents one of the most important conceptual diculties in quantum mechanics.
Consequently, this topic of research has gained considerable interest in the last decades.[2,
3] The presence of a classical apparatus considerably aects the behavior of the observed
quantum system[4, 5, 6, 7] and measurements typically fail to have outcomes of the sort the
theory was created to explain.
Another conceptual diculty is that in a system under observation there are many
degrees of freedom such that information can be lost in the couplings which may account
for dissipation. One possible approach that has often been used to answer this question is
to introduce all degrees of freedom for the bath and solve a number of coupled equations
in various limits of some approximation. In fact, by using the inuence-functional method,
it has been shown[8] that dissipation tends to destroy quantum interference in a time scale
shorter than the relaxation time of the system. This result has given justication for the
use of logarithmic nonlinear wave equations[9]-[47] to describe quantum dissipation. These
equations have been validated as an appropriate, practical bath functional in time-dependent
density functional theory for open quantum systems.[14]
This work addresses both conceptual diculties mentioned above. In particular, an
answer to a challenge posed by Bell on the dividing line between the quantum and classical
regimes in a measurement problem is given here. To this end, a generalized logarithmic
nonlinear Schr odinger equation is proposed to describe the time evolution of a quantum
dissipative system under continuous measurement. Within the Bohmian mechanics frame-
work, a solution to this equation reveals a novel result: it displays a time constant which
establishes the dividing line between the quantum and classical trajectories. As in simple
RC circuits and complex electronic pacemakers, the time constant determines the dividing
line between charging, discharging and normal heartbeats. It is shown that continuous mea-
surements and damping not only disturb the particle but compel the system to converge in
time to a Newtonian regime without any assumption of collapse. While the width of the
wave packet may reach a stationary regime, its quantum trajectories converge exponentially
in time to classical trajectories. In particular, it is shown that damping tends to suppress
further quantum eects on a time scale shorter than the relaxation time of the system. If the
initial wave packet width is taken to be equal to 2.8 10
15
m (the approximate size of an
electron), the Bohmian time constant is found to have an upper limit, i. e.,
B max
= 10
26
s.
Bohmian mechanics has recently attracted increasing attention from researchers.[15]-
[45] Despite the uncertainty principle, the predictions of nonrelativistic quantum mechanics
permit particles to have precise positions at all times. The simplest theory demonstrating
that this is so is indeed Bohmian mechanics. One of the fundamental aspects of Bohmian
mechanics is its ability to tackle more clearly the quantum measurement problem. The
wave function plays a dual role in the Bohmian model; it determines the probability of the
actual location of the particle and monitors its motion. As pointed out by Bell,[1] in physics
the only observations we must consider are position observations - a denite outcome in an
individual measurement is determined by the relevant position variable associated with the
apparatus. It is a great merit of the Bohmian picture to force us to consider this fact.
3
The time evolution of the wave function of the quantum dissipative system (x, t) un-
der continuous measurement can be described in terms of a nonlinear Schr odinger equation.
This equation combines two types of logarithmic nonlinearities: (1) For the description of a
system under continuous measurement, Nassar[3] has recently proposed a Schr odinger-type
equation with the nonlinear logarithmic term i h ln ||
2
, along the lines of the pioneer-
ing work of Mensky[2] and Bialynicki-Birula and Mycielski [46]. A remarkable feature of
this equation is the existence of exact solitonlike solutions of Gaussian shape. Hefter [47]
has given physical grounds for the use of this logarithmic nonlinear equation by applying
it to nuclear physics and obtaining qualitative and quantitative positive results. He argues
that this type of equation can be applied to extended objects such as nucleons and alpha
particles. (2) For the description of quantum dissipative systems, Kostin [9] constructed
a Schr odinger-type equation with the nonlinear logarithmic term (i h/2) ln(/). This
equation has the very interesting property that, at the level of observables, it satises the
dissipative Langevin equation at T = 0. This equation has subsequently been derived by
Skagerstam[11] and Yasue[12] and has found extensive use in many applications.[13] The
Kostin nonlinear logarithmic term has recently been suggested by Yuen-Zhou et al. [14] as
an appropriate, practical bath functional in time-dependent density functional theory for
open quantum systems with unitary propagation. So, by combining both nonlinearities, the
generalized logarithmic nonlinear Schr odinger equation reads
i h
(x, t)
t
= [H(x, t) + i h (W
c
(x, t) + W
f
(x, t))] (x, t), (1)
with
W
c
(x, t) =
_
ln |(x, t)|
2

_
ln |(x, t)|
2
__
(2)
and
W
f
(x, t) =

2
_
ln
(x, t)
(x, t)

_
ln
(x, t)
(x, t)
__
, (3)
where the coecient characterizes the resolution of the continuous measurement and the
coecient represents the friction coecient. The terms in <> arise from the requirement
that the integration of Equation (1) with respect to the variable x under the requirement
that for a particle the expectation value of the energy < E(t) >, dened as
< E(t) >
+
_

(x, t)E(t)(x, t)dx, (4)

must be equal to the expectation values of the kinetic and potential energies through the
Hamiltonian H. [9, 48] For the system studied here, no external potential is assumed (i.e.,
V = 0).
4
Equation (1) has several interesting and unique properties. First of all, it guarantees
the separability of noninteracting subsystems. Other nonlinear modications can intro-
duce interaction between two subsystems even when there are no real forces acting between
them. Second, the stationary states can always be normalized. For other nonlinearities,
stationary solutions have their norms fully determined and after multiplication by a con-
stant they cease to satisfy the equation. And third, the logarithmic nonlinear Schr odinger
equation (1) possesses simple analytic solutions in a number of dimensions especially non-
spreading wave-packet solutions. It is fundamentally dierent from the equations proposed
by Bialynicki-Birula and Mycielski [46] due to (1) the imaginary coecient in front of the
logarithmic terms and (2) the last term
_
ln |(x, t)|
2
_
. Equation (1) also generalizes the
equation proposed by Kostin in order to account for continuous observation.
Equation (1) can now be solved via the Bohmian formalism.[15]-[45] To this end, the
wave function is rst expressed in the polar form:
(x, t) = (x, t) exp(iS(x, t)/ h). (5)
Now, after substitution of Equation (5) into Equation (1), we obtain
i h
_

t
+
i
h
S
t

_
=
=
h
2
2m
_
_
_
_
_

x
2


h
2
_
S
x
_
2
_
_
+
i
h
_
2
S
x

x
+

2
S
x
2
_
_
_
_
i h
_
ln
2
< ln
2
>
_
+ [S S] . (6)
Equation (6) can be separated into real and imaginary parts. By dening the quan-
tum hydrodynamical density , velocity v and quantum potential V
qu
respectively as
(x, t) =
2
(x, t), (7)
v =
1
m
S
x
, (8)
V
qu
=
h
2
2m

x
2
, (9)
5
we reach
v
t
+ v
v
x
+ v =
1
m
V
qu
x
(10)
and

t
+

x
(v) + [ln ln ] = 0. (11)
Equation (10) is an Euler-type equation describing trajectories of a uid particle,
with momentum p = mv, whereas Equation (11) describes the evolution of the quantum uid
density . This density is interpreted as the probability density of a particle being actually
present within a specic region. Such a particle follows a denite space-time trajectory that
is determined by its wave function through an equation of motion in accordance with the
initial position, formulated in a way that is consistent with the Schr odinger time evolution.
An essential and unique feature of the quantum potential is that the force arising from it
is unlike a mechanical force of a wave pushing on a particle with a pressure proportional
to the wave intensity. By assuming that the wave packet is initially centered at x = 0 and
(x, 0) = [2
2
(0)]
1/2
exp [x
2
/2
2
(0)] and vanishes for |x| at any time we may
rewrite
(x, t) = |(x, t)|
2
=
_
2
2
(t)
_
1/2
exp
_

[x x(t)]
2
2
2
(t)
_
, (12)
where (t) is the total width of the Gaussian wave packet.
Now, Equation (12) can be readily used to demonstrate that
+
_

_
[x x(t)]
2
_
(x, t)dx =
2
(t). (13)
Substitution of Equation (12) into Equation (11) yields

t
=
_

+
(x x)

x +
1

3
(x x)
2

_
, (14)
and
(v)
x
=
_

_
+
__

_
(x x) +

x
_ _

(x x)

2
_
, (15)
6
which implies that
v(x, t) =
_

_
(x x) +

x. (16)
Analogously, substitution of Equation (16) into Equation (10) yields
_

(t) + ( 2)

(t) + (
2
)(t)
h
2
4m
2

3
(t)
_
(x x)
1
+ (

x +

x)(x x)
0
= 0,(17)
which implies that

(t) + ( 2)

(t) + (
2
)(t) =
h
2
4m
2

3
(t)
(18)
and

x +

x = 0. (19)
Equations (18) and (19) show that continuous measurement of a quantum dissipative
wave packet gives specic features to its evolution: the appearance of distinct classical and
quantum elements, respectively. This measurement consists of monitoring the position of
the quantum dissipative system and the result is the measured classical trajectory x(t) for t
within a quantum uncertainty (t).
The associated Bohmian trajectories [21, 22, 41, 44, 45] of an evolving i
th
particle of
the ensemble with an initial position x
oi
can be calculated by rst substituting
x
i
(t) = v
i
(x, t) (20)
into Equation (16) to obtain
x
i
(t) = x(t) + x
oi
(t)

o
e
t
. (21)
where
o
= (0) is the initial width.
The position of the center of mass of the wave packet (the classical trajectory) is
represented by x(t) while x
oi
is the initial position of the i
th
individual particle in the Gaussian
ensemble corresponding to the wave function given by Equation (5). Now, Equation (18)
admits analytic Gaussian-shaped soliton-like solutions (Gaussons): for = 0, a stationary
regime can be reached and the width of the wave packet can be related to the resolution of
measurement as follows
=

2
+

2
4
+
h
2
4m
2

4
o
, (22)
7
which means that if an initially free wave packet is kept under a certain continuous measure-
ment, its width may not spread in time. It is worth stressing the fact that the inverse of
h
2m
2
o
is associated with the relative spreading of a free Gaussian wave packet [49]; in other words,
the corresponding spreading velocity is given by
h
2mo
. Thus, the eective friction given by
takes into account the two spreading mechanisms present in this dynamics.
The transition from quantum to classical trajectories can then be dened as the
Bohmian time constant to be
B

1
and Equation (21) can be further simplied to
x
i
(t) = x(t) + x
oi
e
t/
B
. (23)
It follows from Equation (23) that if x
oi
= 0, then the particle follows the Newtonian
trajectory at any time. If, however, x
oi
is positive, then the particles distributed in the
right half of the initial ensemble are accelerated whereas the particles distributed in the
left half of the initial ensemble are decelerated. Nevertheless, there is only a temporary
asymmetry in the Bohmian velocities between any two symmetric particles since the rate
of the asymmetry diminishes with time. After a short time, the distance in position space
shifted by the particles initially lying at positive and negative x
oi

s converges to a constant
value. So, continuous measurements not only disturb the particle but compel it to eventually
converge to a classical position. It is also noticeable that damping tends to suppress further
quantum eects on a time scale shorter than the relaxation time of the system. For a small
friction coecient ( <
h
m
2
o
), the Bohmian time constant can be expressed as

B

2m
2
o
h
_
1
m
2
o
h
_
. (24)
Further, from Equations (9) and (21) we have that the quantum force is given by
F
qu
=
V
qu
x
=

x
_

h
2
8m
4
o
(x x)
2
+
h
2
4m
2
o
_
=
h
2
4m
4
o
x
oi
e
t/
B
. (25)
So, the convergence of the quantum particle trajectories to classical trajectories is
due to the inuence of the measuring apparatus and friction through the quantum force.[50]
The quantum force is directly proportional to the initial position of the i
th
particle and
decays exponentially in time (it drops 63% of its initial value after a time constant
B
).
Likewise, the quantum position x
i
(t) - the initial position of the i
th
individual particle in the
Gaussian ensemble - approaches its classical value. So, friction and continuous observation
of a wave packet may lead to a gradual freezing of the quantum features of the particle.
Finally, if the initial wave packet width for an electron is taken to be equal to
2.810
15
m (the approximate size of an electron [51]) and the coecient of friction is made
very small ( <<
h
m
2
o
), the Bohmian time constant is found to have an upper limit:

B max
= 10
26
s. (26)
8
This result provides an answer to a challenge posed by Bell [1, 42] on the lack of
clarity about the line between the quantum and classical regimes in a measurement problem:
The Bohmian time constant above dened may establish that dividing line.[52]
9
Acknowledgments
Part of this work was done during the Summer 2012 at the UCLA Physics and As-
tronomy Department. Correspondence with I. Bialynicki-Birula is gratefully acknowledged.
S. M-A grateful acknowledges the MICINN (Spain) through Grant FIS2011-29596-C02-01.
10
References
[1] J. S. Bell, Speakable and Unspeakable in Quantum Mechanics (Cambridge University
Press, Cambridge, 1987).
[2] M. B. Mensky, Phys. Lett. A231, 1 (1997); Phys. Lett. A307, 85 (2003); Int. J. Theor.
Phys. 37, 273 (1998);Continuous Quantum Measurement and Path Integrals (IOP Pub.,
Bristol, 1993). See references therein.
[3] A. B. Nassar, Ann. Phys. 331, 317 (2013). See references therein.
[4] D. Home and M. A. B. Whitaker, J. Phys. A: Math. Gen. 25, 657 (1992) and references
therein.
[5] B. Misra and E. C. G. Sudarshan, J. Math. Phys. 18, 756 (1977); C. B. Chin, E. C. G.
Sudarshan and B. Misra, Phys. Rev. D 16, 520 (1977).
[6] A. Peres, Am. J. Phys. 48, 931 (1980); A. Peres and A. Ron, Phys. Rev. A 42, 5720
(1990).
[7] W.M. Itano, D.J. Heinzen, J.J. Bollinger and D.J. Wineland, Phys. Rev. A 41, 2295
(1990).
[8] A. O. Caldeira and A. J. Leggett, it Phys. Rev. A31, 31 (1985). In particular, see their
conclusions.
[9] M. D. Kostin, J. Chem. Phys. 57, 3589 (1972).
[10] J. H. Weiner and R. E. Forman, Phys. Rev B10, 314 (1974).
[11] B. K. Skagerstam, Phys. Lett. 64B, 21 (1975).
[12] K. Yasue, Ann. Phys. (NY) 114, 479 (178).
[13] J. J. Grin and K. K. Kan, Rev. Mod. Phys. 48, 467 (1976).
[14] J. Yuen-Zhou, D. G. Tempel, C. A. Rodriguez-Rosario and A. Aspuru-Guzik, Phys.
Rev. Lett. 104 043001 (2010).
[15] C. L. Lopreore and R. E. Wyatt, Phys. Rev. Lett. 82, 5190 (1999).
[16] C. L. Lopreore and R. E. Wyatt, Chem. Phys. Lett. 325, 73 (2000).
[17] C. C. Chou and R. E. Wyatt, Phys. Rev. Lett. 107, 030401 (2011).
[18] C. C. Chou and R. E. Wyatt, Phys. Rev. A 76, 012115 (2007).
[19] D. D urr, S. Goldstein, R. Tumulka, and N. Zanghi, Phys. Rev. Lett. 93, 090402 (2004).
See references therein.
[20] E. Guay and L. Marchildon, J. Phys. A: Math. Gen. 36, 5617 (2003).
[21] A. S. Sanz, F. Borondo and S. Miret-Artes, Phys. Rev. B61, 7743 (2000).
11
[22] A. S. Sanz and S. Miret-Artes, Am. J. Physics 80, 525 (2012). This works oers a
comprehensive analysis of the current status Bohmian mechanics. See references therein.
[23] R. Guantes, A.S. Sanz, J. Margalef-Roig, S. Miret-Artes, Surface Science Reports 53,
199 (2004).
[24] D. Bohm, Phys. Rev. 85 (1952) 166; Phys. Rev. 85, 180 (1952).
[25] R. Tumulka, Am. J. Phys. 72, 1220 (2004). See references therein.
[26] Oleg V. Prezhdo and Craig Brooksby, Phys. Rev. Lett. 86, 3215 (2001).
[27] Y. Nogamia, F. M. Toyamab and W. van Dijk, Phys. Lett. A270, 279 (2000).
[28] A. R. Plastino, M. Casas and A. Plastino Phys. Lett. A281, 297 (2001).
[29] N. C. Dias and J. N. Prata Phys. Lett. A302, 261 (2002).
[30] A. J. Makowski, , P. Pepowski and S. T. Dembiski Phys. Lett. A266, 241 (2000).
[31] L. Shifren, , R. Akis and D. K. Ferry Phys. Lett. A274, 75 (2000).
[32] P. Falsaperla and G. Fonte Phys. Lett. A316, 382 (2003).
[33] A. Datta, P. Ghose and M. K. Samal Phys. Lett. A322, 277 (2004).
[34] C. Colijn and E. R. Vrscay Phys. Lett. A300, 334 (2002).
[35] G. Potel, M. Muoz-Alear, F. Barranco and E. Vigezzi Phys. Lett. A299, 125 (2002).
[36] R. G. Stomphorst Phys. Lett. A292, 213 (2002).
[37] Md. M. Ali, A. S. Majumdar and D. Home Phys. Lett. A304, 61 (2002).
[38] G. E. Bowman Phys. Lett. A298, 7 (2002).
[39] S. Goldstein, Bohmian mechanics in Stanford Encyclopedia of Philosophy, edited by E.
N. Zalta, http://plato.stanford.edu/archives/win2002/entries/qm-bohm/.
[40] A. B. Nassar, J. Math. Phys. 27, 2949 (1986).
[41] A. K. Pan, Pram. J. of Phys. 74, 867 (2010).
[42] J. Bernstein Am. J. Phys. 79, 601 2011. This work presents a master analysis of Bohmian
mechnaics.
[43] P. Holland, Ann. Phys. 315, 505 (2005).
[44] P. Holland, The Quantum Theory of Motion (Cambridge Univ. Press, 1993). See pages
68-70.
[45] R. E. Wyatt, Quantum Dynamics with Trajectories (Springer, New York, 2005). See
pages 61, 150 and 151.
12
[46] I. Bialynicki-Birula, Brazilian J. of Phys. 35, 211 (2005). It is fundamentally dierent
from the equation proposed by Bialynicki-Birula and Mycielski due to the imaginary
coecient in front the logarithmic term.
[47] E. F. Hefter, Phys. Rev. A32, 1201 (1985).
[48] For a free particle, the expectation value of the total energy must be equal to the
expectation value of the kinetic energy,
i.e., i h
+
_

(x, t)
(x,t)
t
dx =
h
2
2m
+
_

(x, t)

2
(x,t)
x
2
dx or < E(t) >=<
P
2
(t)
2m
> since
+
_

_
[x x(t)]
2

2
(t)
_
|(x, t)|
2
dx = 0 by using Equation(12). See also Equation (13)
[49] A. S. Sanz and S. Miret-Artes, J. Phys. A: Mat. Theor. 41, 435303 (2008).
[50] From Equation (12), we note that the expectation value of the quantum force vanishes
at all times: F
qu
=
1
m
_
Vqu
x
_
=
h
2
2m
2
x x(t) = 0.
[51] Experiments to measure the size of the electron consist on colliding two beams of elec-
trons against each other and counting how many are scattered and altered their trajec-
tories. By counting the collisions, and knowing how many particles we have thrown, we
can estimate the average size of each particle in the beam.
[52] In RC circuits, for example, the time constant is a measure of how quickly a capacitor
becomes charged or discharged. In electronic pacemakers, the time constant regulates
the pulsing rate of the heart and determines the dividing line between normal heartbeats.