1 Concept
The principle of quantum superposition is most clearly
understood when it refers to pure states of a quantum system.
1 CONCEPT
surfaces would not do. Likewise, the constituent micro beams.[8] The output that is found in the interference patscopic particles, such as atoms, of a crystal diractor are tern is not some partial or fractional state such as perhaps
physically arranged with spatial periodicity.
classical thinking might expect. No, it is a pure state that
The ltered output is here called the 'primary beam'. It is either detected or not detected, with a denite probais pure with respect to the lter. This means that if it is bility. The probabilistic occurrence of such pure states is
passed through a copy of the lter, it comes out practically a principle that is characteristic of quantum physics.
intact, not signicantly reduced by a second ltration.[2]
Alternatively, the primary beam can be prepared by special techniques, so as to consist of practically individual,
mutually incoherent, and mutually unentangled photons
or atoms, each associated with a 'herald' quantal entity
that indicates when the object quantal entity is to be detected. Such experiments are less common than those
with partly coherent and entangled beams.
1.2
For perfect superposition it is essential that the intermediate beams are mutually coherent. That is to say, they
are all physically derived from one and the same primary
beam in a pure state. Moreover, for maintenance of coherence, there must be no intrusive factor in the several
intermediate beam paths that aects some of the quantal entities dierently from others. In other words, each
and every one of the quantal entities in the beam must be
exposed to one and the same arrangement of ight paths.
Otherwise, superposition is imperfect.
Analysis of the primary beam into sec It is evident that for this scheme to work, the analyser
arrangement considered as a whole is unaltered by inondary subbeams
terchanging its input and output of the primary beam.
In a manner of speaking, passage of the quantal entities
through the analyser arrangement as a whole is reversible.
This is reected in the Hermitian nature of the mathematical operators, called observables, that represent the
devices such as analysers. In contrast to this, the combination of production of the beam and its destruction by a
detector is irreversible.
The primary pure beam then passes into a beam splitter or measurement device of the rst kind, or
premeasurement device, that will here be called a 'quantum analyser'[3][4][5][6] that has multiple output channels.
All the output channels are kept open, but otherwise the
analyser is in many respects like the lter. Again, it is a
macroscopic object that has mutually entangled quantum
mechanical collective modes. Consequently, the emerg The principle was described by Paul Dirac as follows:
ing secondary subbeams are coherent, perhaps more coherent than they were in the raw beam.
The general principle of superposition of quantum mechanics apThe quanta emerge probabilistically as subbeams in the
plies to the states [undisturbed moanalysers several output channels. Respectively, the
tions] ... of any one dynamical syssubbeams are then in states pure with respect to the
tem. It requires us to assume that
analyser.[7]
between these states there exist peculiar relationships such that whenever the system is denitely in one
1.3 Reconstitution of the primary pure
state we can consider it as being
beam
partly in each of two or more other
states. The original state must be
The subbeams are then passed, through a carefully conregarded as the result of a kind of
trived spatial arrangement, to a copy of the analyser in
superposition of the two or more
a reverse posture, intended to reconstitute the primary
new states, in a way that cannot be
beam.
conceived on classical ideas. Any
The principle of quantum superposition states that,
state may be considered as the reprovided the primary beam is pure, it is possible so to
sult of a superposition of two or
carefully contrive the spatial arrangements, that the remore other states, and indeed in
sult is a perfect restoration of the primary input beam.
an innite number of ways. ConThe primary pure state has been restored. It is said to be
versely any two or more states may
a superposition of the several intermediate pure states.
be superposed to give a new state...
...
If the spatial arrangement is not exactly the one that restores the primary pure state, in general the output of the
The nonclassical nature of the sureconstitutive copy analyser is split or analysed with defperposition process is brought out
inite probabilities into the several output channels. If they
clearly if we consider the superposiare reassembled, but not in the special way that restores
tion of two states, A and B, such that
the original beam, they produce what is called an interthere exists an observation which,
ference pattern. Again it is said to be a superposition,
when made on the system in state
a dierent but denite one, of the several intermediate
A, is certain to lead to one particular
3
result, a say, and when made on the
system in state B is certain to lead to
some dierent result, b say. What
will be the result of the observation
when made on the system in the superposed state? The answer is that
the result will be sometimes a and
sometimes b, according to a probability law depending on the relative weights of A and B in the superposition process. It will never be
dierent from both a and b [i.e, either a or b]. The intermediate character of the state formed by superposition thus expresses itself through
the probability of a particular result
for an observation being intermediate between the corresponding probabilities for the original states, not
through the result itself being intermediate between the corresponding
results for the original states."[9]
1.4
Decoherence
Alternatively to the foregoing case of coherent reassembly of the split beams, if one the several split beams is
not sent on for reassembly but instead is interrupted by
a detector, the detected state is in general dierent from
that of the primary pure beam; it is said to be decohered
from it, because it has not been exposed to the possibility of coherent reassembly.[10] In a manner of speaking,
the reconstitutive second analyser has been replaced by
a lter with a detector in its output channel. The term
'registration' is sometimes used to refer to this.[11][12]
2 Theory
2.1 Examples
For an equation describing a physical phenomenon, the
superposition principle states that a combination of solutions to a linear equation is also a solution of it. When
this is true the equation is said to obey the superposition
principle. Thus if state vectors f 1 , f 2 and f 3 each solve
the linear equation on , then = c1 f 1 + c2 f 2 + c3
f 3 would also be a solution, in which each c is a coecient. The Schroedinger equation is linear, so quantum
mechanics follows this.
For example, consider an electron with two possible congurations, up and down. This describes the physical system of a qubit.
c1  + c2 
is the most general state. But these coecients dictate
probabilities for the system to be in either conguration.
The probability for a specied conguration is given by
the square of the absolute value of the coecient. So the
probabilities should add up to 1. The electron is in one of
those two states for sure.
pup = c1 2
1.5
1.6
pdown = c2 2
pdown or up = pup + pdown = 1
Continuing with this example: If a particle can be in state
up and down, it can also be in a state where it is an amount
3i/5 in up and an amount 4/5 in down.
 =
3
4
i +  .
5
5
2
= 9 . The probIn this, the probability for up is 3i
5
25
4 2
= 16 . Note that 9 + 16 = 1
ability
for
down
is
moving indepen5
25
25
25
.
Isolated particle,
dently, not in a beam of many particles
For an isolated single instance of a quantal entity, considered without respect to any quantum lter or analyser, purity, mixture, and superposition are undened. The single isolated quantal entity is simply what it is in itself. A
classically thinking observer therefore sees no quantum
superposition. For a classical thinker, the fundamental
 
THEORY
A(x, y)x, y
xy
 
4
3i
 + 
5
5
Then after those 10 seconds our state will change to

(
c1  + c2  c1 () + c2
)
4
3i
 + 
5
5
x
In probability theory there is a similar principle. If a sysThe principle of superposition guarantees that there are tem has a probabilistic description, this description gives
states which are arbitrary superpositions of all the posi the probability of any conguration, and given any two
tions with complex coecients:
dierent congurations, there is a state which is partly
this and partly that, with positive real number coecients, the probabilities, which say how much of each
there is.
(x)x
x
+ (x)x, + (x)x,
2.3
Hamiltonian evolution
real and imaginary parts separately, the sign of the coecients is important. In probability, two dierent possible
n n
outcomes always add together, so that if there are more
n
options to get to a point z, the probability always goes up.
that
In quantum mechanics, dierent possibilities can cancel. So
the
innite
list
of
amplitudes
completely describes the quantum
In probability theory with a nite number of states, the
state of the particle. This list is called the state vector,
probabilities can always be multiplied by a positive numand formally it is an element of a Hilbert space, an
ber to make their sum equal to one. For example, if there
innite dimensional complex vector space. It is usual to
is a three state probability system:
represent the state so that the sum of the absolute squares
of the amplitudes add up to one:
(...2 ,1 ,0 ,1 ,2 ...)
n n = 1
where the probabilities x, y, z are positive numbers.
Rescaling x,y,z so that
For a particle described by probability theory random
walking on a line, the analogous thing is the list of probabilities (...P2 , P1 , P0 , P1 , P2 , ...) , which give the
x+y+z =1
probability of any position. The quantities that describe
how they change in time are the transition probabilities
The geometry of the state space is a revealed to be a tri K
xy (t) , which gives the probability that, starting at x,
angle. In general it is a simplex. There are special points the particle ends up at y after time t. The total probain a triangle or simplex corresponding to the corners, and bility of ending up at y is given by the sum over all the
these points are those where one of the probabilities is possibilities
equal to 1 and the others are zero. These are the unique
locations where the position is known with certainty.
well described. The geometry of the phase space can be P (t + dt) = P (t) + dt
Px Rxy
y
y
viewed as a hint that the quantity in quantum mechanx
ics which corresponds to the probability is the absolute
where Rxy is the time derivative of the K matrix:
square of the coecient of the superposition.
A2r + A2i + Br2 + Bi2 + Cr2 + Ci2 = 1
2.3
Hamiltonian evolution
Rxy =
Kxy (dt) xy
dt
The numbers that describe the amplitudes for dierent The equation for the probabilities is a dierential equation
possibilities dene the kinematics, the space of dier which is sometimes called the master equation:
ent states. The dynamics describes how these numbers
change with time. For a particle that can be in any one of
innitely many discrete positions, a particle on a lattice, dPy = P R
x xy
the superposition principle tells you how to make a state: dt
x
THEORY
The R matrix is the probability per unit time for the particle to make a transition from x to y. The condition that
the K matrix elements add up to one becomes the condi (I + iH dt)(I iHdt) = I
tion that the R matrix elements add up to zero:
H H = 0
which says that H is Hermitian. The eigenvalues of the
Hermitian matrix H are real quantities which have a physRxy = 0
ical interpretation as energy levels. If the factor i were aby
sent, the H matrix would be antihermitian and would have
One simple case to study is when the R matrix has an purely imaginary eigenvalues, which is not the traditional
equal probability to go one unit to the left or to the right, way quantum mechanics represents observable quantities
describing a particle which has a constant rate of random like the energy.
walking. In this case Rxy is zero unless y is either x+1,x,
or x1, when y is x+1 or x1, the R matrix has value For a particle which has equal amplitude to move left and
c, and in order for the sum of the R matrix coecients right, the Hermitian matrix H is zero except for nearest
to equal zero, the value of Rxx must be 2c. So the neighbors, where it has the value c. If the coecient is
everywhere constant, the condition that H is Hermitian
probabilities obey the discretized diusion equation:
demands that the amplitude to move to the left is the complex conjugate of the amplitude to move to the right. The
dPx
equation of motion for is the time dierential equation:
= c(Px+1 2Px + Px1 )
dt
Unm
Unp = mp
2
= 2
t
x
U U = I
i
= 2 + V (x)
t
x
d
Umn
dt
7
between them. In a statistical system in discrete time,
t=1,2,3, described by a transition matrix for one time step
d
n =
Hnm m
Kmn , the probability to go between two points after a i
dt
nite number of time steps can be represented as a sum
whose Hamiltonian has the same eigenvalues as those of
over all paths of the probability of taking each path:
the R matrix of the statistical system. The eigenvectors
are the same too, except expressed in the rescaled basis.
Ht
theory, the probability m for the stochastic matrix obeys K (t) = e
detailed balance when the stationary distribution n has
For quantum systems which are invariant under time rethe property:
versal the Hamiltonian can be made real and symmetric,
so that the action of timereversal on the wavefunction
is just complex conjugation. If such a Hamiltonian has
n Knm = m Kmn
a unique lowest energy state with a positive real waveDetailed balance says that the total probability of going function, as it often does for physical reasons, it is confrom m to n in the stationary distribution, which is the nected to a stochastic system in imaginary time. This reprobability of starting at m m times the probability of lationship between stochastic systems and quantum syshopping from m to n, is equal to the probability of go tems sheds much light on supersymmetry.
ing from n to m, so that the total backandforth ow of
probability in equilibrium is zero along any hop. The condition is automatically satised when n=m, so it has the 3 Experiments and applications
same form when written as a condition for the transitionprobability R matrix.
Successful experiments involving superpositions of
relatively large (by the standards of quantum physics) objects have been performed.[13]
n Rnm = m Rmn
A cat state has been achieved with photons.[14]
When the R matrix obeys detailed balance, the scale of
A beryllium ion has been trapped in a superposed
the probabilities can be redened using the stationary disstate.[15]
tribution so that they no longer sum to 1:
pn
= n pn
n Rnm
= Hnm
m
and H is symmetric
Hnm = Hmn
This matrix H denes a quantum mechanical system:
4 FORMAL INTERPRETATION
states of a macroscopic quantal entity. For the
principle of superposition, after it is prepared
but before it is detected, it may be regarded
as exhibiting an intermediate state. It is not
a singleparticle state such as is often considered in discussions of interference, for example
by Dirac in his famous dictum stated above.[19]
Morever, though the 'intermediate' state may
be loosely regarded as such, it has not been
produced as an output of a secondary quantum
analyser that was fed a pure state from a primary analyser, and so this is not an example of
superposition as strictly and narrowly dened.
Nevertheless, after preparation, but before
measurement, such a SQUID state may be regarded in a manner of speaking as a pure
state that is a superposition of a clockwise and
an anticlockwise current state. In a SQUID,
collective electron states can be physically prepared in near isolation, at very low temperatures, so as to result in protected coherent
intermediate states. Remarkable here is that
there are found two wellseparated respectively
selfcoherent collective states that exhibit such
metastability. The crowd of electrons tunnels
back and forth between the clockwise and the
anticlockwise states, as opposed to forming a
single intermediate state in which there is no
denite collective sense of current ow.[20][21]
In contrast, for actual real cats, such wellseparated metastable collective states states do
not exist and consequently cannot be physically prepared. Schrdingers point was that
classical thinking does not in general anticipate such physically distinct and separate
metastable quantum states. In classical thinking, distinct quantum states even of single
atoms can indeed be regarded as metastable,
and are remarkable and unexpected. In the
days when Schrdinger raised his argumentative example, no one had imagined the invention of SQUIDs that exhibit such states on a
macroscopic scale. The presentday physicist
here pays close attention to the requirement
mentioned above, that the intermediate states
must be carefully physically shielded to protect
them from any factor that aects some of the
independent quantal entities (in this case collective not single particle) dierently from others. Contrary to this requirement, the living
cat breathes. This destroys intermediate state
coherence, and so the conditions required for
exhibition of the principle of superposition are
not fullled.
4 Formal interpretation
Applying the superposition principle to a quantum mechanical particle, the congurations of the particle are all
positions, so the superpositions make a complex wave in
space. The coecients of the linear superposition are a
wave which describes the particle as best as is possible,
and whose amplitude interferes according to the Huygens
principle.
For any physical property in quantum mechanics, there is
a list of all the states where that property has some value.
These states are necessarily perpendicular to each other
using the Euclidean notion of perpendicularity which
comes from sumsofsquares length, except that they also
must not be i multiples of each other. This list of perpendicular states has an associated value which is the value of
the physical property. The superposition principle guarantees that any state can be written as a combination of
states of this form with complex coecients.
Write each state with the value q of the physical quantity
as a vector in some basis nq , a list of numbers at each
value of n for the vector which has value q for the physical
quantity. Now form the outer product of the vectors by
multiplying all the vector components and add them with
coecients to make the matrix
Anm =
q
qnq m
A piezoelectric "tuning fork" has been constructed, where the sum extends over all possible values of q. This
which can be placed into a superposition of vibrating matrix is necessarily symmetric because it is formed from
9
the orthogonal states, and has eigenvalues q. The matrix
A is called the observable associated to the physical quantity. It has the property that the eigenvalues and eigenvectors determine the physical quantity and the states which
have denite values for this quantity.
6 See also
Eigenstates
MachZehnder interferometer
Physical interpretation
It is natural to ask why ordinary everyday real (macroscopic, Newtonian) objects and events do not seem empirically to display quantum mechanical features such as
superposition. Indeed, this is sometimes regarded even
as mysterious, for example by Richard Feynman.[26] In
1935, Erwin Schrdinger devised a wellknown thought
experiment, now known as Schrdingers cat, which highlighted the dissonance between quantum mechanics and
Newtonian physics, where only one conguration occurs, although a conguration for a particle in Newtonian
physics species both position and momentum.
The explanation is as follows. It is a logical truism that
a single detection of a quantal entity, observed alone,
empirically considered, is not an example of a relation
of several states. For the several states are not empirically dened when the quantal entity is observed alone.
It would therefore be nonsense to try to say that it, a single
state, observed alone, empirically shows superposition.
Superposition is a relation of several states that are empirically dened only when several intermediate beams
are empirically veried to be present. Actual empirical
observation of superposition requires that the several intermediate beams should be actually observed, in several
distinct experimental setups. Without the several experiments, talk of superposition is mere theoretical speculation, not empirical observation. The superposed state is,
as dened above, also a pure state, with respect to the primary analyser. It is classically inexplicable how a quantum analyser can have several pure states as outputs. That
is Feynmans mystery.
Penrose Interpretation
Pure qubit state
Quantum computation
Schrdingers cat
Wave packet
7 References
[1] CohenTannoudji, C., Diu, B., Lalo, F. (1973/1977).
Quantum Mechanics, translated from the French by S.R.
Hemley, N. Ostrowsky, D. Ostrowsky, second edition,
volume 1, Wiley, New York, ISBN 0471164321, p. 260.
[2] Feynman, R.P., Leighton, R.B., Sands, M. (1965), Chapter 6.
[3] Merzbacher, E. (1961/1970), p. 219.
[4] Bartell, L.S. (1980). Complementarity in the doubleslit
experiment: on simple realizable systems for observing
intermediate particlewave behavior, Phys. Rev. D 21:
16981699.
[5] Aspect, A., Dalibard, J., Roger, G. (1982). Experimental tests of Bells inequalities using timevarying analysers,
Phys. Rev. Lett. 49: 18041807.
[6] McIntyre, D.H. (2012).
Quantum Mechanics: a
Paradigms Approach, Pearson AddisonWesley, San
Francisco, ISBN 9780321765796, Chapter 1.
[7] Feynman, R.P., Leighton, R.B., Sands, M. (1965), Chapter 6.
Quantum superposition is exhibited in fact in many di [8] Feynman, R.P., Leighton, R.B., Sands, M. (1965), Chapter 5.
rectly observable phenomena, such as interference peaks
from an electron wave in a doubleslit experiment. Super[9] Dirac P.A.M. (1930/1958), p. 12.
position persists at all scales, provided that coherence is
shielded from disruption by intermittent external factors. [10] Bacciagaluppi, G. (2012). The role of decoherence in
quantum mechanics.
This is a reason for dierences of opinion, as between the
10
7.1
REFERENCES
Wheeler, J.A.; Zurek, W.H. (1983). Quantum Theory and Measurement. Princeton NJ: Princeton University Press.
11
8.1
Text
8.2
Images
File:Ambox_important.svg Source: http://upload.wikimedia.org/wikipedia/commons/b/b4/Ambox_important.svg License: Public domain Contributors: Own work, based o of Image:Ambox scales.svg Original artist: Dsmurat (talk contribs)
File:Editclear.svg Source: http://upload.wikimedia.org/wikipedia/en/f/f2/Editclear.svg License: Public domain Contributors: The
Tango! Desktop Project. Original artist:
The people from the Tango! project. And according to the metadata in the le, specically: Andreas Nilsson, and Jakub Steiner (although
minimally).
File:Quantum_superposition_of_states_and_decoherence.ogv Source:
http://upload.wikimedia.org/wikipedia/commons/8/8b/
Quantum_superposition_of_states_and_decoherence.ogv License: CC BYSA 3.0 Contributors: Own work Original artist: Jubobro
J.Bobro and full list in credits
File:Wiki_letter_w_cropped.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/1c/Wiki_letter_w_cropped.svg License:
CCBYSA3.0 Contributors:
Wiki_letter_w.svg Original artist: Wiki_letter_w.svg: Jarkko Piiroinen
8.3
Content license