Professional Documents
Culture Documents
PDF Advances in Open Systems and Fundamental Tests of Quantum Mechanics Proceedings of The 684 We Heraeus Seminar Bad Honnef Germany 2 5 December 2018 Bassano Vacchini Ebook Full Chapter
PDF Advances in Open Systems and Fundamental Tests of Quantum Mechanics Proceedings of The 684 We Heraeus Seminar Bad Honnef Germany 2 5 December 2018 Bassano Vacchini Ebook Full Chapter
https://textbookfull.com/product/advances-in-databases-and-
information-systems-22nd-european-conference-adbis-2018-budapest-
hungary-september-2-5-2018-proceedings-andras-benczur/
https://textbookfull.com/product/open-quantum-systems-dynamics-
of-nonclassical-evolution-subhashish-banerjee/
Programming Languages and Systems 16th Asian Symposium
APLAS 2018 Wellington New Zealand December 2 6 2018
Proceedings Sukyoung Ryu
https://textbookfull.com/product/programming-languages-and-
systems-16th-asian-symposium-aplas-2018-wellington-new-zealand-
december-2-6-2018-proceedings-sukyoung-ryu/
https://textbookfull.com/product/quantum-mechanics-a-fundamental-
approach-1st-edition-k-kong-wan/
Springer Proceedings in Physics 237
Bassano Vacchini
Heinz-Peter Breuer
Angelo Bassi Editors
Advances in Open
Systems and
Fundamental
Tests of Quantum
Mechanics
Proceedings of the 684. WE-Heraeus-
Seminar, Bad Honnef, Germany,
2–5 December 2018
Springer Proceedings in Physics
Volume 237
Indexed by Scopus
The series Springer Proceedings in Physics, founded in 1984, is devoted to timely
reports of state-of-the-art developments in physics and related sciences. Typically
based on material presented at conferences, workshops and similar scientific
meetings, volumes published in this series will constitute a comprehensive
up-to-date source of reference on a field or subfield of relevance in contemporary
physics. Proposals must include the following:
– name, place and date of the scientific meeting
– a link to the committees (local organization, international advisors etc.)
– scientific description of the meeting
– list of invited/plenary speakers
– an estimate of the planned proceedings book parameters (number of pages/
articles, requested number of bulk copies, submission deadline).
Angelo Bassi
Editors
123
Editors
Bassano Vacchini Heinz-Peter Breuer
Dipartimento di Fisica Physikalisches Institut
Università degli Studi di Milano Albert-Ludwigs-Universität Freiburg
Milan, Italy Freiburg im Breisgau, Baden-Württemberg, Germany
Angelo Bassi
Department of Physics
University of Trieste
Trieste, Italy
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
v
vi Preface
include the idea to quantify the exchange of information between the open system
and its environment in terms of the distinguishability of quantum states and to
interpret the backflow of information from the environment to the open system as
characteristic feature of quantum non-Markovianity. Another example is the idea to
characterize memory effects of quantum processes in open systems through the
concept of the divisibility of the corresponding quantum dynamical map. Both ideas
also lead to a quantification of the degree of memory effects in open quantum
systems. Experimental realization of non-Markovian quantum systems and mea-
surements of the degree of memory effects have been carried out in both photonic
and trapped ion systems. A further important research topic is the study of the
impact of classical or quantum system-environment correlations. Several general
schemes have been developed theoretically and implemented experimentally which
enable the detection of such correlations by means of only local operations acting
on the open system. Examples of applications are the experimental determination of
initial correlations in photonic systems and the local detection of quantum phase
transitions in trapped ion systems. In addition to these advances in the general
characterisation and classification as well as in the experimental detection and
control of quantum processes in open systems, there has been important conceptual
and technical progress in the analytical and numerical treatment of strongly coupled
non-Markovian systems. Relevant developments have also been put forward in the
microscopic modelling of such systems employing, for example, collision models
to derive large classes of physically admissible quantum master equations.
This volume collects some recent developments in the field of open quantum
systems and foundations of quantum mechanics that were the subject of the 684.
WE-Heraeus-Seminar on Advances in Open Systems and Fundamental Tests of
Quantum Mechanics. The event was generously funded by the Wilhelm und Else
Heraeus-Stiftung and took place in the beautiful environment of the Physikzentrum
in Bad Honnef, Germany, on December 2018.
vii
Contributors
ix
x Contributors
J. F. Haase Institut für Theoretische Physik and IQST, Universität Ulm, Ulm,
Germany
S. F. Huelga Institut für Theoretische Physik and IQST, Universität Ulm, Ulm,
Germany
David Picconi Institute of Physical and Theoretical Chemistry, Goethe University
Frankfurt, Frankfurt am Main, Germany
Ángel Rivas Departamento de Física Teórica, Facultad de Ciencias Físicas,
Universidad Complutense, Madrid, Spain;
CCS-Center for Computational Simulation, Campus de Montegancedo UPM,
Madrid, Spain
A. Smirne Institut für Theoretische Physik and IQST, Universität Ulm, Ulm,
Germany
Roberta Zambrini IFISC (UIB-CSIC), Instituto de Fisica Interdisciplinar y
Sistemas Complejos Universitat de les Illes Balears-Consejo Superior de
Investigaciones Cientificas, Palma, Spain
Chapter 1
Collapse Models: Main Properties
and the State of Art of the Experimental
Tests
1.1 Introduction
Quantum mechanics is the most precise theory we have for describing the micro-
scopic world. However, since its formulation, the theory never stopped to raise issues
regarding its meaning. In particular, the superposition principle does not seem to
apply to the macroscopic world. This raises the well-known measurement problem.
Collapse models provide a phenomenological solution to such a problem. These
models modify the Schrödinger equation by adding stochastic and nonlinear terms,
which implement the collapse of the wave function [1]. An in-built amplifica-
tion mechanism ensures that their action is negligible for microscopic systems and
becomes stronger when their mass increases thus providing a natural implementation
of the quantum-to-classical transition.
The most supported among collapse models are the Ghirardi–Rimini–Weber
(GRW) [2] and the Continuous Spontaneous Localization (CSL) models [3, 4]. Their
M. Carlesso (B)
Department of Physics, University of Trieste, Strada Costiera 11, 34151 Trieste, Italy
e-mail: matteo.carlesso@ts.infn.it
Istituto Nazionale di Fisica Nucleare, Via Valerio 2, 34127 Trieste, Italy
S. Donadi
Frankfurt Institute for Advanced Studies (FIAS), Ruth-Moufang-Strasse 1, 60438
Frankfurt am Main, Germany
L̂ a |ψ
|ψ → , (1.1)
|| L̂ a |ψ||
1 Itpossible to define the model also through a stochastic differential equation describing the inter-
action with a Poissonian noise, see [6, 7].
2 In their original formulation [2], Ghirardi, Rimini, and Weber considered the possibility that
different particles can have different collapse rate λi . However, this is not required and in literature
only one λ, representing the collapse rate for a nucleon, is considered. For composite objects, the
corresponding total collapse rate can be calculated through the amplification mechanism discussed
below.
1 Collapse Models: Main Properties and the State of Art of the Experimental Tests 3
2
− (q̂−a)
L̂ a = (πrC2 )−3/4 e .
2
2rC
(1.2)
Now, let us suppose that the jth particle collapses around a. As in the single-particle
case, this implies that Ψ gets multiplied by a Gaussian centered in a, namely,
( j)
L̂ a |Ψ
|Ψ → ( j)
, (1.5)
|| L̂ a |Ψ ||
4 M. Carlesso and S. Donadi
( j)
with L̂ a = (πrC2 )−3/4 exp −(q̂ j − a)2 /2rC2 . Consequently, the ψ−a branch of the
superposition is suppressed. Since the collapses happen independently for any jth
particle, then the state in (1.4) collapses with an amplified rate Λ = N λ.
In the GRW model, the collapse does not preserve the symmetry of the wave func-
tion implying that the model cannot describe identical particles. This limitation was
overcome with the CSL model, which was formulated using the second quantiza-
tion formalism. Thus, it automatically guarantees that its dynamics preserve the wave
function symmetry. In this model, the collapse is described by a nonlinear interaction
with a classical noise. The CSL equation reads [1]
√
d|ψt λ
i
= − Ĥ + dx M̂(x) − M̂(x) M̂(x) wt (x)
dt m0 t
(1.6)
λ
2
− 2 dx M̂(x) − M̂(x) M̂(x) |ψt ,
2m 0 t
where |ψt is the N particle wave function and Ĥ is the system Hamiltonian. Here
m 0 is a reference mass taken as that of a nucleon, and wt (x) is the noise providing the
collapse, characterized by E[wt (z)] = 0 and E[wt (z)ws (x)] = δ (3) (z − x)δ(t − s),
where E[ · ] denotes the stochastic average over the noise. The locally averaged mass
density operator is defined as
M̂(x) = mj dy g(x − y)â †j (y, s)â j (y, s), (1.7)
j s
where â †j (y, s) and â j (y, s) are, respectively, the creation and annihilation operators
of a particle of type j with spin s at the point y, while
(x−y)2
1 −
(2rC2 )
g(x − y) = 3/2
e , (1.8)
π 3/4 rC
is a smearing function imposing the spatial correlation of the collapses. Exactly as for
the GRW model, also in the CSL model, the wave function gets localized in space.
Indeed, the effect of the second and the third terms in (1.6) is to induce a localization
in the eigenstates of the operators M̂(x) [9], which are position eigenstates. The
mass proportionality of M̂(x) guarantees automatically the implementation of the
amplification mechanism.
Regarding the amplification mechanism, the mass proportionality of M̂(x) auto-
matically implements it. However, in CSL model, the amplification factor is different
1 Collapse Models: Main Properties and the State of Art of the Experimental Tests 5
compared to that in the GRW model. Indeed, in CSL, the amplification factor depends
on the shape of the considered system, and not just on the number N of its nucle-
ons. In the particular case of a rigid body, when its size is smaller than rC , we have
Λ = N 2 λ. Conversely, in the limit of rC smaller that the interparticle distance, the
amplification scales with Λ = N λ, which is the same as in the GRW model.
Working directly with (1.6) is in general problematic, because the equation is
nonlinear. However, as long as we are interested in computing expectation values,
we can replace the CSL dynamics with [9]
√
d|ψt λ
i = Ĥ − dx M̂(x)wt (x) |ψt (1.9)
dt m0
We can divide the possible tests of collapse models in two classes of experiments:
interferometric and non-interferometric ones. Interferometric experiments are the
most natural choice of testing collapse models since they detect the direct action
of collapse models. One prepares a quantum system in a superposition state and
then measures the corresponding interference pattern. The collapse action will be
determined by the reduction of the interference contrast. Figure 1.1 summarizes the
state of the art of the bounds on the collapse parameters inferred from interferometric
experiments, where different bounds are shown: in green and in blue from cold
atoms [10] and molecular [11–14] interferometry, respectively, and in orange from
entanglement experiments with diamonds [15, 16]. By following the same reasoning,
one derives also which is the minimum action that collapse models should impose
to actually solve the measurement problem at the macroscopic level. Specifically,
a lower bound (gray area) is derived by requiring that a superposition of a single-
layered graphene disk of radius 10−5 m collapses in less than 10−2 s [12].
In the second class of possible tests of collapse models, one exploits an indirect effect:
the Brownian-like motion induced by the interaction of the collapse noise with the
considered system. This motion imposes a growth of the position variance of the
center of mass of the system, which can be eventually measured. Alternatively, if
the system is charged, one can measure the radiation emission due to its acceleration
given by such a motion. Since no superposition is involved in these experiments, one
can make use of systems of truly macroscopic dimensions. Indeed, due to the in-built
6 M. Carlesso and S. Donadi
Fig. 1.1 Exclusion plots for the CSL parameters from interferometric experiments with respect
to the GRW’s and Adler’s theoretically proposed values [2, 5]: molecular interferometry [11, 12]
(blue area), atom interferometry [10] (green area), and experiment with entangled diamonds [15,
16] (orange area). We report with the gray area the region excluded from theoretical arguments
[12]. M, d, and T refer, respectively, to the mass, the superposition distance involved and the time
of each experiment
amplification mechanism, the collapse effect becomes stronger and thus easier to be
detected. However, larger systems are also more affected by environmental noises,
which compete with that due to the collapses. Thus, to impose strong bounds on CSL
parameters, one seeks for a large mass in an experiment that should be as noiseless
as possible.
Figure 1.2 summarizes the state of the art in this class of experiments, which
includes experiments involving cold atoms [17, 18], optomechanical systems [19–
28], X-ray measurements [29, 30], and phonon excitations in crystals [31, 32], and
planetary observations [33].
Of particular interest is the nanomechanical cantilever experiment described in
[21], where an excess noise of known origin was detected. Its value is compatible
with that predicted by the CSL model with—up to date—still non-excluded values
of the CSL parameters. Several standard mechanisms, able to describe such excess
noise, were considered and excluded. An eventual identification of such noise to a
standard source will improve the bound of the experiment in [21] of one order of
magnitude in λ, see the two orange upper bounds contouring the top gray area in
Fig. 1.2.
1 Collapse Models: Main Properties and the State of Art of the Experimental Tests 7
Fig. 1.2 Exclusion plots for the CSL parameters from non-interferometric experiments: cold
atoms [18] (orange area); nanomechanical cantilevers [20, 21] (purple and gray-orange bounded
areas); gravitational wave detectors AURIGA, LIGO, and LISA Pathfinder [26, 27, 34] (red, blue,
and green areas, respectively); X-ray measurements [29, 30] (light blue area); and phonon exci-
tations in crystals [31, 32] (red line). Similar bounds can be also found from planetary observa-
tions [33]. M and T refer, respectively, to the mass and the time of the experiment
There are some limitations on the regime of validity of GRW and CSL models. To
make an example, both models are non-relativistic. Possible relativistic extensions
have been suggested for the GRW model in [35] as well as for the CSL model in [36].
Moreover, GRW and CSL models have other two weaknesses. The first is the
presence of a steady increase in the energy of any system in time, the second is the
use of a white (flat) noise. Here, we discuss how such limitations can be evaded.
In the CSL model, the energy of any system is not conserved due to the interaction
with the noise inducing the collapse. In the case of a free single particle, one has [1]
3mλ2
Ĥ t = Ĥ 0 + t. (1.10)
4m 20 rC2
The energy of the system grows indefinitely in time. For example, a hydrogen atom is
heated by 10−14 K per year considering the values λ = 10−16 s−1 and rC = 10−7 m.
Although the increment is small, this feature is not realistic even for a phenomenolog-
ical model. Here, the CSL noise acts as an infinite temperature bath. Conversely, one
8 M. Carlesso and S. Donadi
expects that a system will eventually thermalize at the finite temperature of the noise.
The introduction of dissipation precisely guarantees this. Indeed, in the dissipative
CSL model, (1.10) becomes
Ĥ t = e−χt Ĥ 0 − H as + H as , (1.11)
4km λ 2 2
2
with χ = (1+k) 5 m 2 and H as = 16kmr 2 , where k = 8mk r 2 T
3
. Here, T CSL is a new
0 C B C CSL
parameter representing the effective temperature of the noise. Theoretical arguments
suggest T CSL = 1 K.
For a detailed discussion on the dissipative extension of the CSL (and GRW)
model, the reader can refer to [6, 37]. Here, we give an intuition on how dissipation
is included in the model. Consider the Fourier transform of the localization operators
in the CSL and the dissipative CSL model. They are given, respectively, by
rC2
mj
dP dQ e− Q·y e− 22 Q â †j (P + Q)â j (P),
i 2
M̂(y) = (1.12)
j
(2π )3
and
rC2
mj
dP dQe− Q·y e− 22 |(1+k j )Q+2k j P| â †j (P + Q)â j (P).
i 2
M̂ D (y) =
j
(2π )3
(1.13)
Here, the action of the operator â †j (P + Q)â j (P) is to destroy a particle with momen-
tum P and to create another one with momentum P + Q, i.e., to transfer a momentum
Q to the system. In the CSL model, the distribution of the transferred momentum Q
is a Gaussian centered around zero and it does not depend on the system momentum
P. This is the reason why the noise keeps heating the system indefinitely. On the
contrary, in the dissipative CSL model, the distribution of the possible transferred
momentum is centered around a point proportional to −P. In this way, the energy of
any system approaches an asymptotic finite value in the longtime regime.
Figure 1.3 shows the upper bounds of the dissipative CSL extension for different
values of T CSL . For a more detailed analysis on the current bounds of the dissipative
CSL model, the reader may refer to [12, 14, 18, 38].
The second limitation of the CSL model is that the noise inducing the collapse
is white. While this can be a good approximation in certain regimes, no real
noise is expected to be completely white. In particular, it is reasonable that for
high enough frequencies the spectrum of the noise presents a cutoff Ω C , whose
inverse denotes a characteristic correlation time of the noise. Then, it is important to
verify if the presence of a non-white noise affects the model, in particular whether the
localization and amplification mechanism are still working. A detailed and analysis
1 Collapse Models: Main Properties and the State of Art of the Experimental Tests 9
Fig. 1.3 First and second panels: Upper bounds on the dissipative CSL parameters λ and rC
for two values of the CSL noise temperature: T CSL = 1 K (first panel) and T CSL = 10−7 K (second
panel). Third and fourth panels: Upper bounds on the colored CSL parameters λ and rC for
two values of the frequency cutoff: Ω c = 1015 Hz (third panel) and Ω c = 1 Hz (fourth panel).
Red, blue, and green lines (and respective shaded regions): Upper bounds (and exclusion regions)
from AURIGA, LIGO, and LISA Pathfinder, respectively [26]. Purple region: Upper bound from
cantilever experiment [21]. Orange and gray top regions: Upper bound from cold atom experiment
[17, 18] and from bulk heating experiments [31]. The bottom area shows the excluded region based
on theoretical arguments [12]
for generic collapse equations can be found in [9, 39]. In general, one can prove that
both the aforementioned mechanism work. Regarding the predictions of the model,
one derives a stochastic Schrödinger equation with the same form as (1.9) where
the noise wt (x) is substituted by a noise ξt (x) with zero average and correlation
E[ξt (z)ξs (x)] = δ (3) (z − x) f (t, s). Here, f (t, s) denotes the time correlation func-
tion. Note that, contrary to the white noise case where the equation is exact, when
working with colored noise, (1.9) is given by a first-order expansion in λ. Since the
noise effects are typically small, a perturbative treatment is generally sufficient.
10 M. Carlesso and S. Donadi
Some experiments are more sensible than others when a colored noise is intro-
duced. For example, the predictions from radiation emission are sensibly modified.
Indeed, already a cutoff smaller than the order of ∼ 1021 Hz suppresses the cor-
responding bound [40–42] Bounds on the CSL parameters for colored noise were
studied in detail in [12, 18, 43]. In particular, one finds out that the upper bounds
from experiments at high frequencies (or involving small time scales) are weakened
more and more when moving to smaller value of Ω C . Theoretical arguments suggest
Ω C ∼ 1012 Hz. Figure 1.3 shows the upper bounds to the colored CSL extension for
different values of Ω C .
To confirm or falsify the possibility that the excess noise measured in [21] is actu-
ally the effect of a collapse mechanism, one needs to consider new experimental
techniques for an independent inquiry.
One possible test consists of focusing on the rotational degrees of freedom of
a system and its collapse-induced Brownian motion [34, 44]. It turns out that for
truly macroscopic systems, this technique can provide a sensible improvement of the
bounds on the collapse parameters, cf. Fig. 1.4. A direct application was considered
in [34], where the bound from LISA Pathfinder [26] can be significantly improved
by considering also the rotational degrees of freedom.
Another proposal [45] considered a modification of the cantilever experiment in
[21] where the homogeneous mass is substituted with one made of several layers of
two different materials. This will increment the CSL noise for the values of rC of the
order of the thickness of the layers. An example is shown in Fig. 1.4.
These are just two of the several proposals [46–50] suggested over the last years
to push the exploration of the CSL parameter space.
1.8 Conclusions
Fig. 1.4 Exemplification of two possible experimental tests of collapse models. First panel: Results
of the analysis proposed in [34, 44] where the rotational degrees of freedom of a cylinder are
studied. The red line denotes the upper bound that can be obtained from the constraints given by
the rotational motion, compared with those from the translations (blue and green lines). Second
panel: Red shaded area highlights the hypothetical excluded value of the collapse parameters that
could be to derived from the conversion of the translational noise of LISA Pathfinder to rotational
one [34]. This is compared to the new (old) upper bounds from the translational motion shown
with the blue line [34] (gray area [26]). Third panels: Hypothetical upper bounds obtained from
substituting the sphere attached to the cantilever used in [21] with a multilayer cuboid of the same
mass for various thickness of the layers [45]. The bounds are compared with that from the improved
cantilever experiment [21] shown in orange. Fourth panel: Same as the third panel, but with a mass
ten times larger
Acknowledgements MC acknowledges the financial support from the H2020 FET Project TEQ
(grant n.766900) and the support from the COST Action QTSpace (CA15220), INFN and the Uni-
versity of Trieste. SD acknowledges the financial support from the Fetzer Franklin Foundation and
the support from the COST Action QTSpace (CA15220) and the Frankfurt Institute for Advanced
Studies (FIAS). Both the authors are grateful for the support offered by the WE-Heraeus-Stiftung for
the WE-Heraeus-Seminars entitled “Advances in open systems and fundamental tests of quantum
mechanics”.
12 M. Carlesso and S. Donadi
References
2.1 Introduction
Dynamics of open quantum systems attracts a lot of attention [1–3]. Theory of open
quantum systems provides an effective tool to describe the evolution of a quan-
tum system interacting with environment. One of the key questions posed recently
D. Chruściński
Institute of Physics, Faculty of Physics, Astronomy and Informatics,
Nicolaus Copernicus University, Grudzia̧dzka 5/7, 87–100 Toruń, Poland
e-mail: darch@fizyka.umk.pl
Á. Rivas (B)
Departamento de Física Teórica, Facultad de Ciencias Físicas,
Universidad Complutense, 28040 Madrid, Spain
e-mail: anrivas@ucm.es
CCS-Center for Computational Simulation, Campus de Montegancedo UPM,
28660 Boadilla del Monte, Madrid, Spain
S. Chakraborty
Optics and Quantum Information Group, The Institute of Mathematical Sciences,
C. I. T. Campus, Taramani, Chennai 600113, India
Λt = Vt,s Λs , (2.1)
where Vt,s : T (H) → T (H) is a linear map for every t ≥ s. Moreover, one calls Λt
P-divisible if Vt,s is positive trace-preserving (PTP) for all t ≥ s, and CP-divisible
if Vt,s is completely positive trace-preserving (CPTP) for all t ≥ s.
According to [8], the evolution represented by Λt is Markovian iff the corresponding
dynamical map Λt is CP-divisible. A different approach identifies the presence of
memory effects with an information backflow between the system and environment.
This can be implemented in terms of the function
d
σ (ρ1 , ρ2 , t) = Λt ρ1 − Λt ρ2 1 , (2.2)
dt
which depends on two different initial states ρ1 and ρ2 . Following [9], Markovian
evolution is characterized by σ (ρ1 , ρ2 , t) ≤ 0. Whenever σ (ρ1 , ρ2 , t) > 0, one says
there is information backflow from environment back to the system. In this case,
the evolution displays clearly nontrivial memory effects and is classified as non-
Markovian.
These two notions of quantum non-Markovianity are very much related but not
strictly equivalent [10–14]. In this paper, we carefully analyze the relation between
these approaches.
is well defined on T (H). It should be stressed that even if Λt is invertible for all
t > 0 the inverse needs not be completely positive, but it is always trace-preserving.
Actually, one has the following.
Proposition 2.1 ([17]) Let Φ : T (H) → T (H) be a CPTP map. Then the inverse
Φ −1 is CPTP iff
Φ(X ) = U XU † , (2.4)
Φ(X ) = U X T U † . (2.5)
The connection between information backflow and divisibility properties starts with
the following theorem.
Theorem 2.1 ([10]) Suppose that Λt is invertible for all t > 0. Then it is P-divisible
if and only if
d
Λt (X )1 ≤ 0, (2.6)
dt
for all X † = X .
Then, the proof of Theorem 1 is as follows. One has
d 1
Λt (X )1 = lim Λt+ (X )1 − Λt (X )1
dt →0+
1
= lim Vt+,t Λt (X )1 − Λt (X )1 . (2.8)
→0+
and hence
d
Λt (X )1 ≤ 0. (2.10)
dt
Conversely, if (2.6) is satisfied, then (2.8) implies
18 D. Chruściński et al.
for all Y ∈ T (H) and by Lemma 1 the map Vt+,t is positive which ends the
proof.
This result can be immediately generalized for k-divisible maps [16]. These are
dynamical maps such that idk ⊗ Vt,s is PTP on the space Mk (C) ⊗ T (H), where
Mk (C) is the space of k × k matrices with complex entries, and idk denotes the
identity map on this space. Therefore, 1-divisibility is the same as P-divisibility, and
CP-divisibility means k-divisibility for all k. For finite dimension dim T (H) = d <
∞, Λt is CP-divisible iff it is d-divisible.
Theorem 2.2 ([16]) Suppose that Λt is invertible for all t > 0. Then it is k-divisible
if and only if
d
[idk ⊗ Λt ](X )1 ≤ 0, (2.13)
dt
Theorem 2.3 ([12]) Suppose that Λt is invertible for all t > 0. If dim T (H) = d 2 ,
then it is C P-divisible if and only if
d
[idd+1 ⊗ Λt ](X )1 ≤ 0, (2.14)
dt
This result shows that one may restrict to the original scenario X = ρ1 − ρ2 but the
price one pays is a use of (d + 1)-dimensional ancilla instead of d-dimensional one.
For dynamical maps Λt which are not invertible, the relation between divisibility
properties and information backflow becomes considerably more intricate. Note that
for a non-invertible dynamical map Λt , the existence of Vt,s is not guaranteed for all
pairs t > s. We shall consider here just the finite-dimensional case. Then, one proves
the following proposition.
2 Information Flow Versus Divisibility for Non-invertible Dynamical Maps 19
Vt,s (Y ) := Λt (X ), (2.16)
d
Λt (X )1 ≤ 0, (2.18)
dt
d
[idd ⊗ Λt ](X )1 ≤ 0, (2.20)
dt
for all Hermitian X ∈ T (H) ⊗ T (H), then Λt is divisible and Vt,s : Im(Λs ) →
T (H) defines a completely positive trace-preserving map.
Note that (2.20) implies that Vt,s can be defined on the entire space T (H). However,
Vt,s is CPTP only on Im(Λs ), i.e.,
From the discussion in the previous section, we conclude that, for non-invertible
dynamics, the condition for no information backflow implies the existence of maps
Vt,s which are CPTP on Im(Λs ). The question is whether or not one of such propaga-
tors Vt,s can be chosen to be CPTP on the entire T (H). In other words, the problem
we face is how to extend the propagator Vt,s as a map defined on Im(Λs ) to the entire
space T (H) such that it is CPTP on T (H). This extension problem was already
studied both in mathematics and physics [17–19].
This theorem can not be directly applied in our case since the map Vt,s is not unital
and the subspace Im(Λs ) does not define an operator system. Namely, Im(Λs ) does
not always contain the identity element of T (H). Nevertheless, there is a variant of
this theorem closer to our situation.
This result can be directly applied to our problem since the subspace Im(Λs ) is
spanned by positive operators (density matrices). However, note that Theorem 2.5
does not guarantee that the extension is also trace-preserving.
Another random document with
no related content on Scribd:
that five thousand, belongs to me by rights. Oh, never mind, I’m
going along. Do you hear?” he shouted, his fists raised, “I’m going
along, I tell you. There ain’t a man of you big enough to stop me.
Let’s see you try and stop me going. Let’s see you once, any two of
you.” He filled the barroom with his clamor.
“Lord love you, come along, then,” said the sheriff.
The posse rode out of Keeler that same night. The keeper of the
general merchandise store, from whom Marcus had borrowed a
second pony, had informed them that Cribbens and his partner,
whose description tallied exactly with that given in the notice of
reward, had outfitted at his place with a view to prospecting in the
Panamint hills. The posse trailed them at once to their first camp at
the head of the valley. It was an easy matter. It was only necessary
to inquire of the cowboys and range-riders of the valley if they had
seen and noted the passage of two men, one of whom carried a bird-
cage.
Beyond this first camp the trail was lost, and a week was wasted in
a bootless search around the mine at Gold Gulch, whither it seemed
probable the partners had gone. Then a traveling peddler, who
included Gold Gulch in his route, brought in the news of a wonderful
strike of gold-bearing quartz some ten miles to the south on the
western slope of the range. Two men from Keeler had made a strike,
the peddler had said, and added the curious detail that one of the
men had a canary bird in a cage with him.
The posse made Cribbens’ camp three days after the
unaccountable disappearance of his partner. Their man was gone,
but the narrow hoof-prints of a mule, mixed with those of huge hob-
nailed boots, could be plainly followed in the sand. Here they picked
up the trail and held to it steadily till the point was reached where,
instead of tending southward, it swerved abruptly to the east. The
men could hardly believe their eyes.
“It ain’t reason,” exclaimed the sheriff. “What in thunder is he up
to? This beats me! Cutting out into Death Valley at this time of year!”
“He’s heading for Gold Mountain over in the Armagosa, sure.”
The men decided that this conjecture was true. It was the only
inhabited locality in that direction. A discussion began as to the
further movements of the posse.
“I don’t figure on going into that alkali sink with no eight men and
horses,” declared the sheriff. “One man can’t carry enough water to
take him and his mount across, let alone eight. No, sir. Four couldn’t
do it. No, three couldn’t. We’ve got to make a circuit round the valley
and come up on the other side and head him off at Gold Mountain.
That’s what we got to do, and ride like blazes to do it, too.”
But Marcus protested with all the strength of his lungs against
abandoning the trail now that they had found it. He argued that they
were now but a day and a half behind their man. There was no
possibility of their missing the trail—as distinct in the white alkali as
in snow. They could make a dash into the valley, secure their man,
and return long before their water failed them. He, for one, would not
give up the pursuit, now that they were so close. In the haste of the
departure from Keeler the sheriff had neglected to swear him in. He
was under no orders. He would do as he pleased.
“Go on, then, you darn fool,” answered the sheriff. “We’ll cut on
round the valley, for all that. It’s a gamble he’ll be at Gold Mountain
before you’re half-way across. But if you catch him, here”—he
tossed Marcus a pair of handcuffs—“put ’em on him and bring him
back to Keeler.”
Two days after he had left the posse, and when he was already far
out in the desert, Marcus’s horse gave out. In the fury of his
impatience he had spurred mercilessly forward on the trail, and on
the morning of the third day found that his horse was unable to
move. The joints of his legs seemed locked rigidly. He would go his
own length, stumbling and interfering, then collapse helplessly upon
the ground with a pitiful groan. He was used up.
Marcus believed himself to be close upon McTeague now. The
ashes at his last camp had still been smoldering. Marcus took what
supplies of food and water he could carry, and hurried on. But
McTeague was farther ahead than he had guessed, and by evening
of his third day upon the desert Marcus, raging with thirst, had drunk
his last mouthful of water and had flung away the empty canteen.
“If he ain’t got water with um,” he said to himself, as he pushed on,
“if he ain’t got water with um, I’ll be in a bad way. I will, for a fact.”