0% found this document useful (0 votes)
17 views12 pages

Experimental Implementation of Remote State Preparation by Nuclear Magnetic Resonance

The document experimentally implements remote state preparation of a qubit from one nucleus to another in a molecule over interatomic distances using nuclear magnetic resonance techniques. It achieves full remote state preparation of a special ensemble of qubits with one classical bit of communication according to Pati's scheme.

Uploaded by

Sunny Gupta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
17 views12 pages

Experimental Implementation of Remote State Preparation by Nuclear Magnetic Resonance

The document experimentally implements remote state preparation of a qubit from one nucleus to another in a molecule over interatomic distances using nuclear magnetic resonance techniques. It achieves full remote state preparation of a special ensemble of qubits with one classical bit of communication according to Pati's scheme.

Uploaded by

Sunny Gupta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Experimental Implementation of Remote State Preparation by

Nuclear Magnetic Resonance


Xinhua Peng1 , Xiwen Zhu1 , Ximing Fang2,1, Mang Feng1 , Maili Liu1 , and Kelin Gao1
1 Laboratory of Magnetic Resonance and Molecular Physics,

Wuhan Institute of Physics and Mathematics, the Chinese Academy of Sciences,


arXiv:quant-ph/0202004v1 1 Feb 2002

Wuhan, 430071, People’s Republic of China


2 Department of Physics, Hunan Normal University,

Changsha, 410081, People’s Republic of China

()

Abstract

We have experimentally implemented remote state preparation (RSP) of a

qubit from a hydrogen to a carbon nucleus in molecules of carbon-13 labeled

chloroform 13 CHCl3 over interatomic distances using liquid-state nuclear mag-

netic resonance (NMR) technique. Full RSP of a special ensemble of qubits,

i.e., a qubit chosen from equatorial and polar great circles on a Bloch sphere

with Pati’s scheme, was achieved with one cbit communication. Such a RSP

scheme can be generalized to prepare a large number of qubit states and may

be used in other quantum information processing and quantum computing.

Typeset using REVTEX

∗ Corresponding author. E-mail:xwzhu@[Link]; Fax: 0086-27-87885291.

1
I. INTRODUCTION

Quantum states are the carrier of quantum information. Preparing, controlling, access-

ing and transmitting quantum states constitute the important part of quantum information
theory. However, the non-clone theorem [1] of an unknown quantum state holds back the
effective utilization of quantum information. One of the striking discoveries in this area
is quantum teleportation [2], which transmits an unknown quantum state from the sender

(Alice) to the receiver (Bob) by a maximally entangled channel (ebit) and two classical bit
(cbit) communication, namely 1 ebit+2 cbit teleports 1 qubit. Recently, some interesting
works for remote state preparation (RSP) [7–9], similar to teleportation, have been theo-
retically proposed by Lo [8], Pati [7] and Bennett et al. [9], which can remotely prepare a
quantum state from Alice to Bob with less classical bit communication than teleportation.

In their papers, they showed that for a special ensemble of states (e.g., qubit states chosen
from the equator or the polar of the Bloch sphere) RSP requires only one bit of classical
communication per qubit, exactly half that of teleportation, i.e., 1 ebit + 1 cbit transmits
1 qubit. But hitherto, though quantum teleportation has been implemented experimentally

[3–5], RSP has not been experimentally tested yet.


In this paper, we will experimentally demonstrate RSP for a special quantum state (i.e.,
a qubit chosen from equator or polar great circles of a Bloch sphere), with a 2-qubit sample,
carbon-13 labeled chloroform 13 CHCl3 , and complete the task that 1 ebit + 1 cbit transmits

1 qubit. For the general case, Lo [8] conjectured that the classical communication cost of
RSP is equal to that of teleportation, and their procedure would be similar. Bennett et al.
[9] showed that, in the presence of a large amount of prior entanglement, the asymptotic
classical communication cost of RSP for general states is one bit per qubit. However, the
experimental verification of the asymptotic character requires preparation of a large amount

of entanglement source, which would be difficult to implement by the existing techniques.

2
II. REMOTE STATE PREPARATION OF A SPECIAL ENSEMBLE AND THE

NMR REALIZATION

Consider an arbitrary single qubit state

|ψi = α|0i + β|1i, (1)

where α = cos(θ/2) is chosen to be real and β = sin(θ/2) exp(iφ), a complex number. The

state can be represented by a point in a Bloch sphere with two real parameters θ and φ, and
obtained by a transformation R on |0i with
 
−iφ
 cos(θ/2) − sin(θ/2)e 
R=


 (2)

sin(θ/2)e cos(θ/2)

The task of RSP is that Alice helps Bob in a distant laboratory to prepare the state |ψi
known to her but unknown to Bob.
Pati [7] proposed a procedure for RSP of a qubit chosen from equatorial or polar line on
a Bloch sphere. The schematic circuit for RSP is shown in Fig. 1(a).

Similar to teleportation, at first, Alice and Bob share a two-qubit maximal entangled
state, i.e. an EPR state,

1
|ψ − iAB = √ (|0iA |1iB − |1iA |0iB ). (3)
2

|ψ − iAB can be expanded in different orthogonal basis. We choose the qubit orthogonal basis
{|ψi, |ψ⊥ i},

|ψ − iAB = √1 (|ψiA |ψ⊥ iB − |ψ⊥ iA |ψiB ),


2
(4)
= √1 (|ψiA U −1 |ψiB − |ψ⊥ iA |ψiB ).
2

where |ψ⊥ i = α|1i − β ∗ |0i, β ∗ is the complex conjugate of β, and U −1 is the inverse of U
which is a unitary operator to convert |ψ⊥ i into |ψi. It is possible for Alice to perform a

single particle Von-Neumann measurement under the basis because of the state |ψi known
to her. Conditional on Alice’s two measurement outcomes |ψi and |ψ⊥ i, Bob’s state is

3
either |ψi or U −1 |ψi. Thus, after Alice performs the measurement, she sends the result (one
classical bit information) to Bob, who can then recover the desired state |ψi by deciding
to do nothing E or a unitary operation U according to the received classical information.

For a special ensemble, U can be an affirmative unitary operator to transform |ψ⊥ i into
|ψi. For example, if Alice choose to prepare a polar great circles on a Bloch sphere, i. e.,
φ = 0, |ψi = cos(θ/2)|0i + sin(θ/2)|1i, then U = iσy . Alternatively, if Alice wish to help Bob
prepare an arbitrary equatorial state such as |ψi = √1 (|0i
2
+ eiφ |1i), then U = σz . However,

for a general state (both θ and φ are arbitrary), there is no universal transformation U to
take |ψ⊥ i → |ψi by this protocol. Therefore, RSP of an arbitrary state can be made with
the success probability of 50%.
The network of RSP for a special ensemble shown in Fig. 1(b) corresponds to the

schematic circuit in Fig.1(a), including three steps.


(1) Preparation of the EPR pair |ψ − iAB . In NMR experiments, suppose that Alice
and Bob have spin A and spin B, respectively, consisting of a bipartite spin quantum system
in a 4 × 4 Hilbert space HA ⊗ HB . The EPR pair can be prepared from a pure state |0iA |0iB

by the operations of NOT (Ni flips the sign of spin i), the Hadamard gate (Hi transform the
states according to (|0i → √1 (|0i + |1i), |1i → √1 (|0i − |1i)) and controlled-NOT (CNĀB
2 2

flips spin B if and only if A is |0i ) gates. The NA , HA and CNĀB gates were respectively
   
realized by the pulse sequences XA (π), ȲA ( π2 )X̄A (π) and ȲB ( π2 )ZB π
2
ZA π
2
JAB (π) (pulses
applied from right to left). By simplification methodology [12], the EPR pair was realized

by using the pulse sequence

π π π π
     
ȲB ( )JAB (π)YB X̄B ȲA (π) XA . (5)
2 2 2 2

(2) The single particle Von-Neumann measurement under the qubit basis

{|ψi, |ψ⊥ i}. We use a two-part procedure inspired by Brassard et al. to perform the mea-
surement [10]. Part one of the procedure is to rotate from the basis {|ψi, |ψ⊥ i} into the
computational basis {|0i, |1i} by the transformation R+ = R−1 , inverse of R. Part two of
the procedure is to perform a projective measurement in the computational basis.

4
+
Applying RA on the EPR pair, one gets

+ − 1
RA |ψ iAB → √ (|0iA |ψ⊥ iB − |1iA |ψiB ), (6)
2

For a polar state, the operation


 

+
 cos(θ/2) sin(θ/2) 
RA (θ, φ = 0) = 

,
 (7)
− sin(θ/2) cos(θ/2)

which can be realized by the NMR pulse sequence ȲA (θ) . For an equatorial state, the

operation
 
−iφ
1 1 e
+
(θ = 90◦ , φ) = √ 

RA  (8)
2  −eiφ 1 

which can be realized by the NMR pulse sequence XA (θ1 ) ȲA (θ2 ) XA (θ1 ) , where θ1 =
 √ 
tan−1 (sin φ) , and θ2 = 2 sin−1 cos φ/ 2 .
It is possible to simulate directly the second step in NMR exploiting magnetic gradient
pulses [11]. However, it can be seen from Eq. (6) that, the final state of Alice is one of
|0i and |1i, corresponding to the two different measurement results. Owing to the weak

measurement of NMR, we can employ the conditional unitary operation S, instead of the
projective measurement in the computational basis and the post-measurement operation.
(3) Conditional unitary operation S. It can be verified from Eq. (6), that when
Alice measures the outcome of spin A |0i, Bob performs the transformation U on spin B;
when |1i, Bob does nothing, then two measurement results lead to Bob in |ψi. For a polar

state, U = iσy , the conditional unitary operation S = iσyB E+A + E−A , is implemented by the
pulse sequence

π π π
     
ȲB XB JAB (π) X̄B . (9)
2 2 2

For an equatorial state, U = σz , the conditional unitary operation S = σzB E+A + E−A , is

implemented by the pulse sequence

π π π π π π
           
ȲA X̄A YA ȲB XB YB JAB (π) . (10)
2 2 2 2 2 2

5
(4) The measurement of the final state. After tracing out spin A the state of spin B
is reduced to the expected form |ψi , which indicates the success of RSP. In principle, spin
A could be traced out experimentally by applying a decoupling field during observation of

spin B. But this approach was found to lead to unacceptable sample heating, with resultant
shifts in resonance frequencies. Thus, the tracing out process was implemented in software
by integrating the entire multiplet of spin B after adjusting the right phase.

III. THE NMR EXPERIMENT AND RESULTS

The RSP scheme stated above was implemented by liquid-state NMR spectroscopy with
carbon-13 labeled chloroform 13 CHCl3 (Cambridge Isotope Laboratories, Inc.). To perform
RSP we make use of the hydrogen nucleus (1 H) as the sender (Alice) and the carbon nuclei
(13 C) as the receiver (Bob) in the experiments, transmitting the state from 1 H to 13
C.

Spectra were recorded on a BrukerARX500 spectrometer with a probe tuned at 125.77MHz


for 13 C, and at 500.13MHz for the 1 H . The spin-spin coupling constant J between 13
C and
1
H is 214.95Hz. The relaxation times were measured to be T1 = 4.8sec and T2 = 0.2sec for
the proton, and T1 = 17.2sec and T2 = 0.35sec for carbon nuclei.

The pseudo-pure state was prepared in our experiment using line-selective pulses and
the gradient-pulse techniques [13]. We performed two separate sets of experiments of the
RSP process (shown in Fig.2) from 1 H to 13
C. As all NMR observables are traceless and
the constant item has no effect on the NMR signal, any point on the Bloch sphere can be

expressed as the product form Ix sin θ cos φ + Iy sin θ sin φ + Iz cos θ (apart from a constant
unit matrix), which corresponds to the observable NMR signal. Using Alice’s knowledge
of the qubit, an advanced operation R+ was performed on 1 H for a given set of θ and φ
values. In the first set, a total of 25 qubit states chosen from a range of points on the polar
line of the Bloch sphere (φ = 0) with a θ spacing of π/12 were studied. The second set of

experiments was executed for RSP of 17 equatorial states on the Bloch sphere (θ = π/2) with
a φ spacing of π/8. For each state the experimental spectra of 13 C were recorded from the 13 C

6
nucleus, and their real and imaginary components are plotted in Fig. 2. The experimental
results clearly show the expected line, sine and cosine modulations. It can be seen from
Fig. 2, there exists an same initial phase offset in the fitting curves, which maybe caused

by the systemic errors, such as static magnetic field and rf field inhomogeneities and the
imperfect calibrations of rf pulses, etc.. Furthermore, we reconstructed the density matrices
of the obtained states by quantum state tomography [14], and the maximal relative errors
△ < 18%, where
( )
||ρtheory − ρexp t ||
△ = max , (11)
all states ||ρtheory ||
and ρtheory is the expectative state of RSP, and ρexp t is the reconstructed state by the exper-
imental data. Besides static magnetic field and rf field inhomogeneities and the imperfect

calibrations of rf pulses, the errors in experiments arise from decoherence due to the T2 re-
laxations of nuclei and the environmental effects such as temperature. Nevertheless, it turns
out from Fig. 2 and the reconstructed density matrices that the RSP network is effective
for a special ensemble.

IV. CONCLUSION

In summary, we have experimentally implemented the RSP protocol by using NMR


quantum logic gates and circuits in quantum information. In our experiments, RSP of
a special ensemble was successfully completed by a maximally entangled channel with one

classical bit communication, exactly half that of teleportation, namely 1 cbit → 1 qubit. The
special states distributes on the equatorial and polar circles on the Bloch sphere. The RSP
protocol can be generalized to remotely prepare a large number of general qubit states in the
high-entanglement limit at an asymptotic cost of one bit per qubit [9]. Like teleportation,

RSP can be applied not only to pure states, but also to parts of entangled states which has
been studied by Bennett et al. [9]. Although the communication happens between spins in
angstrom distance, the concept and method of quantum information transmission should be
useful for quantum computation and quantum communication.

7
ACKNOWLEDGMENTS

We thank C. P. Sun for bringing the topic of RSP to our attention and Xiaodong Yang,
Hanzheng Yuan, Xu Zhang and Guang Lu for help in the course of experiments.

8
REFERENCES

[1] W. K. Wootters and W. H. Zurek, Nature 299, 802 (1982).

[2] C. H. Bennett, G. Brassard, [Link], R. Jozsa, A. Peres, and W. K. Wootters, Phys.


Rev. Lett. 70, 1895 (1993).

[3] M. A. Nielsen, E. Knill and R. Laflamme, Nature 396, 52 (1998).

[4] D. Bouwmeester et al. Nature 390, 575 (1997).

[5] D. Boschi, S. Branca, F. De Martini, L. Hardy and S. Popescu, Phys. Rev. Lett. 80,

1121 (1998).

[6] X. Fang, X. Zhu, M. Feng, X. Mao and F. Du, Phys. Rev. A (2000).

[7] Arun K. Pati, Phys. Rev. A 63, 014302 (2000).

[8] Hoi-Kwong Lo, Phys. Rev. A 62, 012313 (2000).

[9] C. H. Bennett, D. P. DiVincenzo, J. A. Smolin, B. M. Terhal and W. K. Wootters,


Phys. Rev. Lett. 87, 077902 (2001).

[10] G. Brassard, S. Braunstein and R. Cleve, Phys. D 120, 43 (1998).

[11] G. Teklemariam, E. M. Fortunato, M. A. Pravia, T. F. Havel and D. G. Cory, Phys.


Rev. Lett. 86, 5845 (2001).

[12] R. Ernst, G. Bodenhausen and A. Wokaun, Principles of Nuclear Magnetic Resonance


in One and Two Dimensions (Oxford Univ. Press, Oxford, 1990).

[13] X. Peng, X. Zhu, X. Fang, M. Feng, K. Gao, X. Yang and M. Liu, Chem. Phys. Lett.
340, 509 (2001).

[14] I. L. Chuang, N. Gershenfeld, M. Kubinec and D. Leung, Proc. Roy. Soc. Lond A 454

447 (1998).

[15] B. Schumacher, Phys. Rev. A 54, 2614 (1996).

9
Figure Captions

Fig. 1 Schematic protocol for RSP proposed by Pati (a) and the corresponding network
of implementing RSP (b). H represents the Hadamard gate and conditional operation on a

spin being in the |1i state and the |0i state are represented by a filled circle and an empty
circle, respectively. R+ is demonstrated in the text.
Fig. 2 Experimental results for RSP of qubits chosen from (a) the polar line, and (b) the
equatorial circle, on the Bloch sphere. Data points and ∗ denote the real and imaginary

parts of the NMR signals from 13 C, respectively. The fitting expectative curves are depicted
with the solid lines.

10
data |ψ >

Alice
measure in the
source |0>
different basis
entangle
Bob target |0> conditional
|ψ >
unitary U 0,U1

(a)

+
Alice |0> N H R

Bob |0> S measure

(b)

Fig. 1
(a)

0.5

−0.5

−1

0 π/4 π/2 3π/4 π 5π/4 3π/2 7π/4 2π

(b)

0.5

−0.5

−1

0 π/4 π/2 3π/4 π 5π/4 3π/2 7π/4 2π

Fig. 2

You might also like