Detection of -partite entanglement and -nonseparability of multipartite quantum states
Yan Hong
School of Mathematics and Science, Hebei GEO University, Shijiazhuang 050031, China
Ting Gao
[email protected]School of Mathematical Sciences, Hebei Normal University, Shijiazhuang 050024, China
Fengli Yan
[email protected]College of Physics, Hebei Normal University, Shijiazhuang 050024, China
Abstract
Identifying the -partite entanglement and -nonseparability of general -partite quantum states are fundamental issues in quantum information theory. By use of computable inequalities of nonlinear operators, we present some simple and powerful -partite entanglement and -nonseparability criteria that works very well and allow for a simple and inexpensive test for the whole hierarchy of -partite entanglement and -separability of -partite systems with running from down to 2. We illustrate their strengths by considering several examples in which our criteria perform better than other known detection criteria. We are able to detect -partite entanglement and -nonseparabilty of multipartite systems which have previously not been identified.
In addition, our results can be implemented in today’s experiments.
pacs:
03.65.Ud, 03.67.-a
I Introduction
Quantum entanglement is an important physical resource in quantum computation and quantum information processing and is also a key feature that distinguishes quantum theory from classical theory. So, it is very valuable for studying the characterization and detection of entanglement for general multipartite systems.
There are two different ways to characterize the entanglement of multipartite systems RPMK09 ; Guhne2009 , one is according to the question ”How many partitions are separable?”, the other is ”How many particles are entangled”. The former is described by -separability, the latter by -partite entanglement. -separability provides a fine graduation of states according
to their degrees of separability, while -partite entanglement provides the hierarchic classification of states according to their degrees of entanglement. Detecting and characterizing -partite entanglement provide a refined insight in entanglement dynamics.
For -partite quantum systems, the -partite entanglement and -nonseparability are two different concepts of multipartite entanglement except that -partite entanglement is equivalent to -nonseparability and genuine -partite entanglement is equivalent to -nonseparability. Both -nonseparability and
-partite entanglement can be used to characterize multipartite entanglement. There are
some approaches to identify genuine -partite entanglement such as based on spin-squeezing inequality VitaglianoHyllus11 and elements of density matrix GuhneSeevinck10 ; HuberMintert10 ; GanHong10 ; GanHong11 ; WuKampermann12 ; ChenMa12 , Bell-type inequalities Seevinck02 , semidefinite program and state extensions Doherty02 ; Doherty04 ; Doherty05 , covariance matrices Gittsovich05 , etc.
The -partite entanglement of -partite quantum systems has been studied by using different tools and some progress has been made.
Some of the complete set of generalized spin squeezing inequalities employed to detect -particle
entanglement and bound entanglement VitaglianoHyllus11 . A hierarchic classification of all states from -partite entanglement to -partite entanglement based on Wigner-Yanase skew information was presented ChenZQ2011 . Entanglement criteria in terms of the quantum Fisher information Hyllus2012 ; Goth2012 were applied to detect several classes of -partite entanglement. Further approach had led to -particle entanglement criteria which was developed by a comparison of the Fisher information and the sum of variances of all local operators Gessner16 .
In recent years, -nonseparability of -partite quantum states has attracted more and more attention and extensively explored in various ways GaoYan14 ; GaoHong13 ; HongGaoYan12 ; ABMarcus10 ; Ananth15 ; HuberLlobet13 ; HongLuoSong15 ; HongLuo16 ; KlocklHuber15 . Some practical convenient inequalities only involving elements of density matrix for detecting -nonseparability were constructed in GaoYan14 ; GaoHong13 ; HongGaoYan12 ; ABMarcus10 ; Ananth15 , which can be used to distinguish -1 different classes of any high dimensional -partite -inseparable states.
In HuberLlobet13 a method that derived from some inequalities in the form of entropy vector was developed to test -nonseparability, whose essence is also elements of density matrix.
-nonseparable states can also be checked by quantum Fisher information and local uncertainty relations HongLuoSong15 ; HongLuo16 .
From the Bloch vector decomposition of a density matrix, the tensor elements can reflect many important characteristics of multipartite quantum states. The correlation tensor norms used for determining -nonseparability in multipartite states KlocklHuber15 .
Much effort has been put into the detection and
characterization of -nonseparability and -partite entanglement in multipartite systems, but it
is far from perfect because of the extremely complex structure of entanglement of multipartite quantum states RPMK09 ; Guhne2009 .
This paper is devoted to giving powerful inequalities that provide, upon violation, experimentally accessible sufficient conditions for -nonseparability and -partite entanglement in -partite systems. Compared with previous works, the resulting criteria are stronger. The inequalities for -nonseparability detection given in Ananth15 can be seen as special case of our results.
In this paper, we will further study the structure of entanglement of multipartite quantum states and present some inequalities to detect -partite entanglement and -nonseparability of multipartite quantum states.
In Sec. II, some concepts and symbols are introduced.
In Sec. III and Sec. IV, we derive a full hierarchy of -partite entanglement criteria and -nonseparability criteria based on linear local operators to characterize multipartite entanglement.
Finally, we show their application ability and point out that they can be adapted in present-day experiments.
II Definitions
The -partite entanglement and -nonseparability are two different concepts involving the partitions of subsystem in -partite quantum system.
An -partite quantum pure state is said to be -producible if it can be written as a tensor product
such that each is a state of at most particles Guhne2009 .
A pure state contains -partite entanglement if it is not (-1)-producible.
For an -partite mixed state , if it can be written as a convex combination of -producible pure states, then it is called -producible.
The individual pure states composing a -producible mixed state may be -producible under different partitions.
If is not -producible, we say that contains -partite entanglement. Here .
An -partite quantum pure state is called -separable if there is a splitting of particles into partitions such that Guhne2009 .
For an -partite mixed state , if it can be written as a convex combination of -separable pure states, then it is called -separable.
If is not -separable, we say that is -nonseparable. Here .
In particular, the -producible (or -separable) states are just fully separable states, and the quantum states being not -producible (or 2-separable) are genuinely entangled states.
For the following statement, let’s introduce some symbols.
Let be the global permutation operator performing simultaneous permutations on all subsystems in ,
while be local permutation operators permuting the two copies of all subsystems contained in the set , that is,
III Detection of -partite entanglement
Now we state our main results in the detection of -partite entanglement.
Theorem 1. If is a -producible quantum state of -partite quantum system , where dim , then
(1)
where for , for , with and being any fully separable states of -partite quantum system, and consists of all possible nonempty proper subsets of . That is, an -partite state does not satisfy inequality (1), then contains -partite entanglement.
Proof. Let us first prove that inequality (1) holds for any -producible pure state. Suppose that pure state with is -producible, where every vector involves at most parties. Then we have
which implies that inequality (1) holds for any -producible pure state.
Next, we prove inequality (1) also holds for any -producible mixed state. Suppose that is -producible mixed state, where is -producible. Then by triangle inequality and the Cauchy-Schwarz inequality, one has
The proof is complete.
Theorem 2. Suppose that is an -partite density
matrix acting on Hilbert space with dim . Let , where and are fully separable states of . If is a -producible, then
(4)
for , and
(5)
for .
Here is a quantum state set.
Of course, if an -partite state does not satisfy the above inequality (4) (respectively, inequality (5)), then contains -partite entanglement () (respectively, 2-partite entanglement).
It should be pointed that there are inequalities in eq.(5)), and violation of any one of them implies 2-partite entanglement.
Proof. To establish the validity of ineq. (4) and ineq. (5) for all
-producible states , let us first verify that this is true
for any -producible pure state .
Suppose that pure state is -producible under partition , and . Then we have
(6)
in case of in same part, and
(7)
in case of in different parts ( with
) where . Moreover, (7) indicates that inequality (5) hols with equality for any 1-producible pure states.
Hence, inequality (4) holds for any -producible pure state.
Now, let us prove (4) also holds for any -producible mixed state. Suppose that is -producible mixed state, where is -producible. Then by triangle inequality and the Cauchy-Schwarz inequality, one has
as desired.
Next, we prove that (5) also holds for any 1-producible mixed state.
Suppose that is -producible mixed state, where is -producible. Then by (7), triangle inequality and the Cauchy-Schwarz inequality, one has
Hence, inequality (5) holds for any 1-producible states. This complete the proof.
IV Detection of -nonseparability
Here we present our main results in the detection -nonseparability.
Theorem 3. If is a -separable -partite quantum state acting on Hilbert space , where dim , then
(9)
Here with and being any fully separable states of -partite quantum system, and consists of all possible nonempty proper subsets of .
Of course, is a -nonseparable -partite state if it violates
the above inequality (9).
The proof is similar to that of Theorem 1.
Criterion 1 of Ref. Ananth15 is the special case of Theorem 3 when .
Theorem 4. If is a -separable -partite quantum state acting on Hilbert space , where dim , then
(10)
for . Inequality (5) holds for any -separable states.
Here ,
with and .
If an -partite state does not satisfy the above inequality
(10), then is not k-separable (k-nonseparable).
The proof is similar to that of Theorem 2.
We need to emphasize that inequality (5) holds for any -separable states because -separable states are the same as 1-producible states in -partite quantum system.
Criterion 2 of Ref. Ananth15 is the special case of Theorem 4 when and .
V Illustration
In this section, we illustrate our main results with some explicit examples. We indeed detects -partite entangled states and -nonseparable
mixed multipartite states which beyond all previously studied criteria.
For convenience of comparison, we first list the criteria which can identify -partite entanglement and -separability of some quantum states.
For an -qubit quantum state , let quantum Fisher information , where with acting on the -th qubit.
(I) If
then is not -producible and contains -partite entanglement Hyllus2012 . Here is the largest integer smaller than or equal
to .
Suppose that is an partite state acting on Hilbert space with dim. Let be the SU() generators and , where means that acts on the -th subsystem.
(III) If an partite state acting on Hilbert space satisfies
then is not -producible and contains 3-partite entanglement VitaglianoHyllus11 .
Here .
(IV) If is a -separable -partite quantum state acting on Hilbert space , where dim . Then Ananth15
Example 1. Consider the -qubit mixed states,
where
We choose for our Theorem 1 and Theorem 3.
The parameter ranges of 10-qubit mixed states containing -partite entanglement (with =3 and = 4) detected by Theorem 1 and (I) are illustrated in Fig. 1. We can see that the quantum states containing 4-partite entanglement in the area enclosed by
red line , the axis, blue line and axis are detected only by Theorem 1.
Similarly, we find that the quantum states containing 5-partite entanglement in the area enclosed by
orange line , the axis, green line and axis are detected only by Theorem 1.
For 10-qubit mixed states , the parameter ranges for -nonseparability (with =3 and = 4) detected by Theorem 3 and (II) are illustrated in Fig. 2. We can see that these -nonseparable quantum states in the area enclosed by
purple line , axis, cyan line , line , cyan line and axis are detected only by Theorem 3. Similarly, we find that these -nonseparable quantum states in the area enclosed by
magenta line , axis, brown line , line , brown line and axis are detected only by Theorem 3.
Figure 1: The 4-partite entanglement and 5-partite entanglement detection for
.
Here the red line represents the threshold of the detection for 4-partite entangled states given by Theorem 1, and
the area enclosed by
red line , axis, line and axis corresponds to the
quantum states containing 4-partite entanglement detected by Theorem 1;
the blue line represents the threshold given by (I),
the area enclosed
by blue line , axis, line and axis corresponds to
the
quantum states containing 4-partite entanglement detected by (I). These quantum states containing 4-partite entanglement in the
area enclosed by red line , axis, blue line and axis are
detected only by Theorem 1. Similarly, the orange line and green line represent the
thresholds of the detection for 5-partite entangled states identified
by Theorem 1 and (I), respectively.
these quantum states containing 5-partite entanglement in the
area enclosed by orange line , axis, green line and axis are
detected only by Theorem 1.
Figure 2: The detection power of Theorem 3 and (II) for
when and .
Here the purple line represents the threshold given by Theorem 3, and
the area enclosed by
purple line , axis, line and axis corresponds to 3-nonseparable
quantum states detected by Theorem 3, while
the cyan lines and represent the thresholds given by (II), and
the area enclosed
by cyan line , axis and line and the area enclosed
by cyan line , axis and line corresponds to
3-nonseparable
quantum states detected by (II). These 3-nonseparable quantum states in the
area enclosed by purple line , axis, cyan line , line , cyan line and axis are only
detected by Theorem 3, but not by (II). Similarly, the magenta line represents the threshold of the detection for 4-nonseparable states identified by Theorem 3, the brown lines and represent the thresholds given by (II), and
the
area encircled by magenta line , axis, brown line , line , brown line and axis contains 4-nonseparable quantum states
detected only by Theorem 3, but not by (II).
Example 2. Let us consider the family of 4-qutrit quantum states,
where and , , .
We choose , and for our Theorem 2 and Theorem 4 .
For 4-qutrit mixed states , Fig. 3 shows that Theorem 2 is more powerful than (III) for the detection of 3-partite entanglement. These quantum states containing 3-partite entanglement in the
area enclosed by red line , axis, blue line and axis only are
detected by Theorem 2, but not by (III).
Figure 3: The 3-partite entanglement detection for
.
The red line represents the threshold given by Theorem 2, and
the area enclosed by
red line , axis, line and axis corresponds to the
quantum states containing 3-partite entanglement detected by Theorem 2.
The blue line represents the threshold given by (III),
and
the area enclosed by
blue line , axis, line and axis corresponds to the
quantum states containing 3-partite entanglement detected by (II).
The area enclosed by red line , axis, blue line and axis are quantum states containing 3-partite entanglement
detected only by Theorem 2, but not by (III).
For these 4-qutrit mixed states , Theorem 4 is also stronger than (IV), which will be illuminated in Fig. 4.
(IV) can determine 3-nonseparability in the area enclosed by green line , axis, and line ,
and Theorem 4 with at the same time.
But Theorem 4 can identify other 3-nonseparable quantum states enclosed by orange line , line and axis with .
In other words, these 3-nonseparable states in the
area enclosed by green line , orange line , and axis are
detected only by Theorem 4, not by (IV).
Figure 4: The detection of 3-nonseparability for
.
The green line represents the threshold given by Theorem 4 with and , and (IV) at the same time.
The orange line represents the threshold given by Theorem 4 with and .
The area enclosed by
green line , axis, and line corresponds to the 3-nonseparable
states detected by Theorem 4 and (IV), the area enclosed by
orange line , line and axis corresponds to the 3-nonseparable
states detected by Theorem 4. Hence, the area enclosed by green line , orange line and axis corresponds to the
3-nonseparable
states detected only by Theorem 4, not by (IV).
VI experimental implementation
Our criteria can be implemented in today’s experiment. Next we provide the local observables required to implement our criteria by using the methods GanHong10 ; GaoHong13 ; GuhneLu2007 ; SeevinckUffink2008 .
The left-hand sides of inequality (1) and (9) can be implemented by two local observables and
because of and ,
where and .
Let , , and
where . Then they satisfy , and .
The right-hand side of inequality (1) and (9) can be implemented by local observables
.
The left-hand sides of inequality (4) and (10) can be implemented by two local observables and
because of and ,
where
Here ,
.
Hence, The left-hand sides of inequality (4) and (10) can be implemented by local observables.
The right-hand sides of inequality (4) and (10) can be implemented by the local observables and .
VII Conclusions
In conclusion, we exploit some nonlinear operators to develop a series of inequalities, which can be used to efficiently identify -partite entanglement and -nonseparability of -partite mixed quantum states in arbitrary dimensional systems. These inequalities present sufficient conditions for the detection of -partite entanglement and -nonseparability, distinguish different classes of multipartite inseparable
states, and can identify some -partite entanglement and -nonseparability that had not been identified so far.
These reveal the practicability and efficiency of our results. Moreover, our criteria can be applied to experiment and we give the corresponding local observables required to implement them.
ACKNOWLEDGMENTS
This work was supported by the National Natural Science
Foundation of China under Grant Nos. 12071110 and 11701135, the Hebei Natural Science
Foundation of China under Grant No. A2020205014 and No. A2017403025, the education Department of Hebei Province Natural Science Foundation under Grant No. ZD2020167, and
the Foundation of Hebei GEO University under Grant No. BQ201615.
References
(1) O. Ghne and G. Tth, Phys. Rep. 474, 1 (2009).
(2) R. Horodecki, P. Horodecki, M. Horodecki, and K. Horodecki, Rev. Mod. Phys. 81, 865 (2009).
(3) G. Vitagliano, P. Hyllus, I. L. Egusquiza, and G. Tth, Phys.
Rev. Lett. 107, 240502 (2011).
(4) O. Ghne and M. Seevinck, New J. Phys. 12, 053002 (2010).
(5) M. Huber, F. Mintert, A. Gabriel, and B. C. Hiesmayr, Phys. Rev. Lett. 104, 210501 (2010).
(6) T. Gao and Y. Hong, Phys. Rev. A 82, 062113 (2010).
(7) T. Gao and Y. Hong, Eur. Phys. J. D 61, 765 (2011).
(8) J. Y. Wu, H. Kampermann, and D. Bru, Phys. Rev. A 86, 022319 (2012).
(9) Z. H. Chen, Z. H. Ma, J. L. Chen, and S. Severini, Phys. Rev. A
85, 062320 (2012).
(10) M. Seevinck and G. Svetlichny, Phys. Rev. Lett. 89, 060401 (2002).
(11) A. C. Doherty, P. A. Parrilo, and F. M. Spedalieri, Phys. Rev. Lett. 88, 187904 (2002).
(12) A. C. Doherty, P. A. Parrilo, and F. M. Spedalieri, Phys. Rev. A 69, 022308 (2004).
(13) A. C. Doherty, P. A. Parrilo, and F. M. Spedalieri, Phys. Rev. A 71, 032333 (2005).
(14) O. Gittsovich, P. Hyllus, and O. Ghne, Phys. Rev. A 82, 032306 (2010).
(15) Z. Q. Chen. Phys. Rev. A 71, 052302 (2005).
(16) P. Hyllus, W. Laskowski, R. Krischek, C. Schwemmer, W. Wieczorek, H. Weinfurter, L. Pezz, and A. Smerzi, Phys. Rev. A 85, 022321 (2012).
(17)G. Tth, Phys. Rev. A 85, 022322 (2012).
(18) M. Gessner, L. Pezz, and A. Smerzi, Phys. Rev. A 94, 020101 (2016).
(19) T. Gao, F. L. Yan, and S. J. van Enk, Phys. Rev. Lett. 112, 180501 (2014).
(20) Y. Hong, T. Gao, and F. L. Yan, Phys. Rev. A 86,
062323 (2012).
(21) T. Gao, Y. Hong, Y. Lu, and F. L. Yan, Europhys. Lett. 104, 20007 (2013).
(22) A. Gabriel, B. C. Hiesmayr, and M. Huber, Quantum Inf. Comput. 10, 829 (2010).
(23) N. Ananth, V. K. Chandrasekar, and M. Senthilvelan, Eur. Phys. J. D 69, 56 (2015).
(24) M. Huber, M. Perarnau-Llobet, and J. I. de Vicente, Phys. Rev. A 88, 042328 (2013).
(25) Y. Hong, S. Luo, and H. Song, Phys. Rev. A 91, 042313 (2015).
(26) Y. Hong and S. Luo, Phys. Rev. A 93, 042310 (2016).
(27) C. Klckl and M. Huber, Phys. Rev. A 91, 042339 (2015).
(28) O. Ghne, C. Y. Lu, W. B. Gao, and J. W. Pan, Phys. Rev. A 76, 030305(R) (2007).
(29) M. Seevinck and J. Uffink, Phys. Rev. A 78, 032101 (2008).