Introduction to Quantum Mechanics:

Till the end of nineteenth century the laws of physics were mostly goverened by classical physics. More precisely one could talk in terms of Newtons Laws of motion which is for a single particle r r r d 2r m 2 = F (r ) dt and for many particles r r r d 2 ri m 2 = Fi ( r1,K, rn ) dt We need to solve these equations to completely determine the trajectory ( all subsequent positions and velocities) of a particle or a set of partilcles r r & (0) ). This provides a when its initial position and velocity are given ( r (0), r

completely determined trajectory in the phase space.

As we know this very simple looking mathematical equation (second order differential equation) is immensely powerful and can explain the behavior of a dust particle as well heavenly bodies with a large number of intermediate things. When electromagnetism was discovered this basic mathematical structure and the physical concept associated with it remained unchanged. Only now r the force F is being electromagnetic in nature the above set of a equations needs to be solved with this electromagnetic force. Behavior of sound , light (electromagnetic wave ) none had challenged this basic mathematical framework. However, the experimental techniques to observe and analyze nature by making measurements on it has been constantly improving and at the beginning of the earlier century people were at a stage when they could probe the matter at atomic scale (Think of Rutherford s experiment).

Analysis of results of such experiments challenged the above Newtonian framework of describing the dynamics of a particle. For example, the planetary model of atom ( say hydrogen atom) which correctly identifies the interaction between the outer side of the atoms ( later known as electrons) with its heavy core ( nucleus) as Coulomb interaction, when tried to explain the stability of such atomic structure which in turn implies the stability of matter, produced unphysical result. However it was already known that such structures must be stable. Moreover the ground state (lowest energy state ) should be E g = , but it turns out to be 13.6ev. When such atoms are excited , they are kicked at higher orbits with higher energy , and then eventually radiates and falls but again finally at E g = 13.6ev . By measuring the frequencies of such radiation one can also conclude that the allowed orbits can only be discrete since their energies 13.6 occur only as 2 ev . n Black body radiation: The theoretical expression for the emission and absorption spectrum of a black-body that combines the Maxwells electromagnetic theory with statistical mechanics (essentially laws of statistics applied to explain the behavior of a collection of large number of particles or waves) shows contradiction with the experimental results on such spectrum. Particularly it turned out, that the anomaly resulted from the assumption that the blackbody spectrum being al electromagnetic wave can continuously exchange energy with its environment. The anomaly was removed by Max Planck (also the beginning of the Quantum Mechanics) when he proposed that such exchange of energy only takes places in discrete quantum [ h (h ) ] Photoelectric effect: Around the same time it was observed that it is also possible to release electrons from a metal by irradiating the metal with light (electromagnetic wave). However classical theory of light could not explain under what condition such electron emission will take place. This was explained by Einstein assuming that light is absorbed by metal only in quanta called photon which was also consistent with Plancks hypothesis.

A large number of such interesting phenomena at the atomic levels started appearing. Though each one them had started receiving some explanation ( and each of these explanations contain a very important departure from the classical theory), it had been felt very strongly that, a common theoretical framework that would replace our Newtonian way of describing the dynamics, was very much required. As can be easily guessed that such a challenging construction took a long circuitous route with a large number of trial and error, finally reaching our present state of understanding ( as available in the present day textbooks in Quantum Mechanics). Instead of going through this full history that we would present quantum mechanics as an alternative formalism that has been there for last several decades and explained a large number of experimental facts. This alternative frame-work was primarily due to effort of Dirac, Schroedinger, Heisenberg and so many other very distinguished physicists . The theoretical framework stands on certain postulates known as fundamental postulates of Quantum Mechanics. Before mentioning these postulates we would provide certain motivations which may be logically related to such postulates. Problem with measurement: Consider we have a beam of light passing through a crystal of tourmaline, which has the property of letting through only light plane polarized perpendicular to its optical axis. Classical electrodynamics tells us what will happen for any polarization of the incident beam. If the beam is polarized perpendicular to the optical axis, it will all go through the crystal; if parallel to the axis none of it will go through; while if polarized at an angle to the axis, fraction sin 2 will go through. How are we to understand these results on a photon basis . A beam of plane polarized light in a particular direction can be considered to be made up of photons each plane polarized in that direction. So if the beam is polarized perperndicular or parallel to the incident axis, we merely have to suppose that each photon will either pass completely, or to be blocked completely as it passes through the crystal. A problem however occurs when the beam is obliquely polarized. Each of the incident photons is then obliquely polarized. It is not clear what will happen to such a photon when it reaches the torumaline. A question about what will happen to a particular photon is not really precise. If one makes such experiment, according to quantum mechanics, sometimes one will find a whole photon with energy equal to the energy of the incident

photon on the back side and the other time one will find nothing. If one repeats the experiment a large number of time, one will find the photon on the back side in a fraction sin 2 of the total number of times. Thus the result of the experiment is not determined according to the classical theory, by the conditions under the control of the experimentalist. The most that can be predicted is a set of possible results with a probability for occurrence of each. The further description provided by quantum mechanics runs as follows. It is supposed that a photon polarized obliquely to the optic axis may be regarded as being partly in the state of polarization parallel to the axis may be regarded as being partly in the state of polarization parallel to the axis and partly in the state of polarization perpendicular to the axis. The state of the oblique polarization may be considered as the result of some kind of superposition process applied to the two states of parallel and perpendicular polarization. This implies a certain special kind of relationship between the various states of polarization, very similar to the relation between polarized beam in classical wave optics, but now applied to the individual photons state of polarization instead of the whole beam. When we make the photon meet a tourmaline crystal, we are subjecting it to a observation. The effect of making this observation is to make force the photon entirely into the state of parallel or entirely into the state of perpendicular polarization. It has to make a sudden jump from being partly in each of these two states to being entirely in one or other of them. Which of the two states it will jump into cannot be predicted, but is governed only by the law of probability.

Particles and Waves: In classical physics particles and waves are two distinct entities. Particles are localized bundles of energies and momentums and their trajectories are defined by specifying their position and momentum each time. A wave in contrast, is a disturbance spread over space and is described by a wave r r function (r , t ) which characterizes the disturbance at the point r at the r time t . For example in the case of the sound wave (r , t ) is the excess air r pressure above the normal and in the case of electromagnetic wave (r , t ) is

the electric field vector . Such wave functions also follow certain equation r of motions such that if we know (r ,0) , we can find it all subsequent times r , namely (r , t ) . Of particular interest to us are waves called plane waves , those are periodic in space and time 2 2 ( x, t ) = A exp i x t = A exp(i ) T Waves as we know, exhibit a phenomenon called interference which is peculiar to them as it is not described by classical mechanics. For example we can think of interference of monochromatic light coming out of two sources, which is the famous Youngs double slit experiment

In case of the light wave, the resultant amplitude is given by adding the amplitudes coming out from two sources as two vectors . Like,

r r r A i ( kx t ) A i ( kx t ) E= e 1 + e 2 , here x 2 = x1 + d sin x1 x2 r r A Thus E = e i ( kx1 t ) (1 + e i ) , where = d sin x1

Now the intensity is given by the square of the modulus of this resultant electric field and hence equal to

I E 2 (1 + e i ) 2 = 4 cos 2

Thus I = 4 I 0 if = 2n , and , I = 0 if, = (2n + 1)

If the same experiment is done with particles, then the intensity at each place on the screen will be just equal to the scalar addition of intensity from the two sources, since the particles do not interfere. It turns out that if we fire mono-energetic electrons, from an electron gun through a double slit, it produces an interference pattern like light. And the same behavior can be logically expected from other subatomic entities like neutron, proton, and even for atoms and molecules themselves. Thus we can conclude electrons are not particles since in that case they are not supposed to interfere and therefore their dynamics cannot be described by Newtons law. Also for a beam of electrons ( coming out of the electron gun), one should r be able to write a wavefunction (r , t ) , such that the intensity of such beam I | | 2 number. However it is much more trickier to determine the behavior of a single electron. For example if the intensity of the electron beam is so low, that at one time there is a single electron, then what does stands for? In that case each electron will be found to come within either the slit 1 and slit 2 and to fell in a random location on the screen. Therefore, it is incorrect to say that one single electron is like a wave which will spread over both the slits to pass through it to reach ultimately the screen to form an interference pattern. Instead the position on which it falls is unpredictable even though all the electrons are prepared in identical state. But nevertheless when a large number of electrons fall through the slits on the screen, it forms an interference pattern. The recipe that is therefore is being suggested is to associate a waver fucntion (r , t ) with each electron modulus square of which will give the probability of r r r Finding that electron at time t at the position r . Thus Pr ob(r , t ) =| (r , t ) | 2 r with the normalization condition | (r , t ) | 2 d 3 r = 1 . The later condition implies that the electron must be found somewhere.

The important thing to remember about this wavefunction is, that, it is not r described by giving an r at every t . That would have made it similar like the Newtonian description. Even if we prepare all the electrons in the same

initial condition and they move under identical condition they end up at r various r s . Therefore the STATE of the electron at any instant cannot r be given by r (t ) ( which is mathematically one, two or three numbers), but by r a complex valued function giving the probability amplitude (r , t ) . Various experiments ( the above two are just prototype examples) forcesd us to accept this peculiar definition of the STATE of the electron. The formalism that will be now developed based on this definition of electron state can explain the discrete nature of atomic spectrum, photoelectric effect, almost the entire quantum chemistry, various phenomenon of nuclear and solid-state physics, various phenomenon in Quantum Optics. Its victory march is continuing till today with successful explanations of various new phenomena such as Quantum Hall Effect, Energy spectrum of quantum dots, various phenomenon at mesoscopic and nanscopic scale, Bose Einstein condensation of cold atoms etc. Building Blocks: In classical mechanics two important things are position and momentum (or velocity) which can be determined using Newtons Law. To keep our discussion simplified let us think of one-dimensional world. We shall also considered all such states at a given time, henceforth omitting t Position: If the state of a given electron is given ( x) , then its position is uncertain. Thus there is an uncertainty associated with its x . However according to the standard theory of probability we can define mean position (or standard deviation about that mean position). We know the that the probability density associated with the state of that electron is given by P ( x) =| ( x) | 2 The mean position is therefore

x =

P( x) xdx =

( x) x ( x)dx

Similarly one can define the standard deviation and the higher moments for the position which is a random variable here. Momentum: 1. As we have point out earlier wave function that is know as plane wave are of special interest. Instead of going into the physical reason for that, let us

tell, a major reason for this is the theorem by Fourier ( revise your knowledge about Fourier Series and transform) that all regular functions can be expressed as a linear combinations of such plane waves. So to begin with we can think of associating with each electron a plane wavefunction , namely ( x) = e ikx . This wavefunction has the interesting property that it gives equal probability of finding the electron anywhere in the space. Since P ( x) =| ( x) | 2 = 1 . This means also the location is completely uncertain. What about the momentum of such electron?

a proof that electron has a momentum p = hk which is well defined and has no uncertainty. That this is indeed the momentum of anything whose state can be given by a plane wave will be proved at later stage once we shall lay down the basic principles. But this already provide us a glimpse if the fact that position and momentum are related in a very nontrivial way in Quantum Mechanics. What will happen about an electron whose e ikx For example, consider, = a1e ikx1 + a 2 e ik 2 x One can immediately see that extracting the information of the momentum from this wave function in a way similar to the previous case will not work. We shall interpret this state to correspond to two possible p states with momentum p1 = hk1 and p 2 = hk 2 with probability | a1 | 2 or | a 2 | 2 . We can immiditaely see that we cannot interpret this wavefunction with a definite | a | 2 p1 + | a 2 | 2 p 2 momentum given by the average 1 = p3 , like the previous case | a1 | 2 + | a 2 | 2

2 . k Therefore according to De Broglies hypothesis, the associated momentum with such probability wave is given by p = hk . We must note this is in no way
Now the wavelength associated with such plane wave is given by =

, since in that case it will have a wave function e ik3 x , hk 3 = p3 , which again has completely uncertain position but with a well defined momentum.

Following this principle and invoking our knowledge in fourier transformation, one can write for any well-behaved wave-function ( x) that can be

associated with an electron, or any other similar particle, like photon, proton, 1 neutron etc. ( x) = a (k )e ikx dk . We may recall that this is just the 2 Fourier transform of the wave-function ( x) . Therefore, . Generalizing the above argument of superposition of 2 two plane waves ( addition of two complex valued function) , the probability of finding such a particle in a state with momentum hk is given by P(hk ) | a (k ) | 2

a (k ) =

dx( x)e

ikx

All these have provide us enough hints that quantum states add like vectors. We can think of the wave function ( x) as a vector in an infinite dimensional vector space. For example in our usual three dimensional vector space a r vector A can be thought of a collection of three numbers ( a1 , a 2 , a3 ), such ) ) ) ) . Here i that A = a1i + a 2 j + a3 k , j , k formed an orthonormal basis. While our r usual three dimensional vector A , needs three numbers to specify it , ( x)

requires a complex number at each x . Therefore it is an infinite dimensional space. Like in our usual vector space a linear combination of two ( or more) r r vectors, namely mA + nB is also a vector, a linear combination of two (or more ) wave function, that is m ( x) + n ( x) is also a wave function. We can now extend this analogy further, and study different properties of these infinite dimensional vector space of ( x)' s in the light of well know properties of the three dimensional vector space. For example, dot product of two such vectors can be also written from the analogy r r A B = Ai Bi < ( x) | ( x) >= dx , which is a complex number
i =1, 2 , 3

The norm of such vector can be given by < | >= dx , which is a real number, since this will give the number of electrons, photons etc.

There can be many orthonormal basis. For example for one dimensional wave function an orthonormal basis will be 1 e ikx , where as for three dimensional wave function the orthonormal k = 2h rr 1 ik r basis is kr = e . To see why such basis are orthonormal, let us note 3 that

(2h) 2

r r r r r r r 1 1 i ( k k ' ) r 3 e d r = 3 3 (k k ' ) = 3 ( p p' ) , any two different 3 3 h (2 ) h =k i =0 basis vectors are orthogonal , same as i j= jk

kr | k ' =

And normality comes from the fact r 3 3 r d p' ( p p' ) = 1 r +i = 1, i .k n | e m = 1 j+i analogus to i e
n

r Any (r ) =

1 (2 ) 2
3

r ikrrr r a ( k )e dk =

1 (2h) 2
3

r r i r dp( p)e h

r p r

r With ( p ) is given by inverse Fourier transform. r r r r r | ' = dk a (k )a' (k ) analogous to A B = Ai Bi

i

Bra and Ket Vctors: At this stage we can introduce the concept of Bra and Ket vectors which can aptly describe quantum mechanical state of any entity such as electron and photon. Such definitions are somewhat abstract. At a lager stage we shall show how the wave function ( x) can be thought of as a particular representation of such bra or ket vector. We have just now learnt that quantum mechanical states behave like vectors. Such vectors can have infinite number of components such as the continuous variables like positions on the screen that will indicate the random positions of the falling electrons in the double slit experiment. It can also be a set of discrete variables like the direction of polarization of a photon. Thus it is desirable to have a special name for describing the vectors which are connected with the states of a system in quantum mechanics. Dirac named them as ket vectors and they are denoted by the symbol | . Such ket vectors can be multiplied by complex numbers and can be added together to give other ket vectors, namely c1 | A + c 2 | B =| R Each state of a dynamical system at a particular time corresponds to a ket vector , the correspondence being such that if a state results from the superposition of certain other states ,its corresponding ket vector is

expressible linearly in terms of the corresponding ket vectors of the other states. In the theory of linear vector space if we introduce a set of vectors , we can always introduce a second set of vectors called the dual vectors. In this case the dual vectors are called bra vectors. The necessity of such dual vectors is the following. The ket vectors that we have introduced above are abstract quantities. Now suppose we associate with each such ket vector some number in terms of which we can express any measurable quantity associated with that state. Then the number can be thought as the scalar product of that ket vector with some new vector. There will be one of these new vectors for each linear function of a ket vetors. The rule of linear vector space consists of the ket vectors will therefore be applicable to the vector space consists of such new vectors. These new vectors will be called bra vectors. We shall now see that one convenient way of thinking these bra and ket vectors as column and row matrices. To make the relation more direct with our ordinary vectors let us write the ket vectors representing a quantum mechanical state with infinite number of possibilities such as spatial location of an electron in the double slit experiment in the following way. Lets denote the state of electron on the screen as f ( x1 ) 0 0 1 0 f ( x ) 1 0 0 0 2 f(xi)|xi | f = f ( x3 ) =f(x1) 0 +f(x2) 0 +f(x3) 1 M + L + f(xn) 0 = i M M M M M 0 0 0 1 f ( x n )

f(x1) x1 x2 x3

f(xi) xi

f(xn) xn

Functions as infinite dimensional vectors

f(xi) = xi | f and xi | x j = ij This is very similar to our ordinary three dimensional vectors 0 0 1 A1 = A ,| i + A2 | A = A1i j + A3 k 2 = 0, | j = 1, | k = 0, i | A = A1 , j | A = A2 , k | A = A3 1 0 0 A3

The above discussion can be summarized by telling that prototype mathematical functions can be thought as generalized vectors defined in an infinite dimensional vector space. The relation between the abstract quantum mechanical ket state | or | f and the function ( x) or f(x) is also now clear. They are the representation of this abstract vectors in a definite set of basis states which are in the above example co-ordinate basis. Once the basis states are defined we can easily find the components of the abstract vector in that basis by taking the inner product with the basis states. Once we obtain these inner products which define the components we can write the abstract vector in terms of the components or namely in terms of a function. Thus in most of the quantum mechanical text books the abstract state | and its co-ordinate space representation ( x) both are used in the same sense to imply a quantum mechanical state. A naive analogue of the representation of an abstract vector in a given basis can also be found in the ordinary three dimensional vector algebra. An abstract vector can be thought just as an arrow drawn in a three dimensional space. The different representations can now be obtained in different coordinate systems such as Cartesian, spherical polar, cylindrical polar etc. In each such co-ordinate system the components along three orthogonal directions are different from each other and they can be obtained by taking the inner product in the same way. However the generalization of the concept of this finite dimensional vector to an infinite dimensional vector space faces an apparent problem when the inner products are defined. This problem was solved by Dirac by introducing the famous Dirac delta function.

Linear Operators:

After coming to terms with the idea that the state of a quantum particle has properties very similar to the vectors, we will now learn what changes one such state into another. This must be very important to understand the dynamics of a quantum mechanical particle, since this has to deal with the evolution of a state under certain condition. Physical observables ( the quantities that we like to measure, like position, momentum, angular momentum, energy etc.) are represented by Hermitian linear operators acting on the wave function ( x) .
a r 1 Linear operator. Let us consider three dimensional vector A = a 2 , and the a 3

following 3 by 3 matrix M=

m11 m21 m31

m12 m22 m32

m13 m23 m33

m a + m12 a 2 + m13 a 3 r r 11 1 We have M ij A j = Bi = m 21 a1 + m 22 a 2 + m23 a3 , which can be written as m a + m a + m a 32 2 33 3 31 1 | B = M | A

The operator M is linear since it satisfies M | A1 + B1 >= M | A1 + M | B1 where , are complex numbers .

i } , Also, C | B = C | M | A is a matrix element. Thus in any basis {e j | M | e i . Also, two such operators generally does not commute. M ij = e
M 1M 2 M 2 M 1

In the same way, linear operator exist for the quantum states , which are function ( D imensional vectors). For example, M =

, since M | ( x) = = ( x) . x x

T he matrix element in this case 1 ( x) | M | 2 ( x) = 1 ( x) In case of our three dimensional ordinary vector space r r i , B =| B = b j | e j and ( M + ) ij = m ji * A =| A = ai | e

2 ( x)dx x

i | M | e j = ai b j mij Thus, A | M | B = ai*b j e

* + i = b B | M + | A = b * |e j ai e j | M j a i mij = A | M | B

Since an observable is represented by a hermitian operator, let us learn the definition of hermitian operator ( matrix). M + is the hermitian conjugate of r r M if for all vectors AandB B | M | A = A | M + | B . M is hermitian if M + = M or mij = m * ji . This is also known as self adjoint. Similarly in the space of wave functions, an operator is hermitian if 1 | | 2 >= 1 2 = 2 | | 1 > = 2 1 . Using this definition it can be easily checked that dimensional function space, but i Similarly

is a Hermitian. (Tutorial problem) x

is not hermitian in 1 x

2 = i i is hermitian . Operators can be anything and need 2 x x x not be a differential operator. For example x itself is an operator ( since acting on a given ( x) it gives another function ( x) = x ( x) . This operator

is also hermitian as can be checked easily. Similarly a number is also another example of a hermitian operator. As we have pointed out already, all physical observable in quantum mechanics are associated with hermitian operators. Here is a useful list Position x x Momentum p (1d) ih x r r Momentum p (3d) i h

p2 2m r Angular momentum L Hamiltonian (energy) H

Kinetic energy

h 2 2 h 2 2 ( 1 d ), (3d ) 2m 2mx 2 r r r ih r h2 2 + V (r ) 2m

Eigenvalues and eigenfunctions . Generally an operator acting on a state vector changes it into another vector. But if the effect of action of an operator on a state vector or wave

function is just to multiply that state or wave function by a constant, we call that state as an eigenstate of that operator and that constant as an eigenvalue corresponding to that eigenstate. For operator

A | = a | . For example take the case of the momentum operator, The

eigenvalue equation is ih ( x) = a ( x) ( x) = e ikx & a = hk . For every k , there is a different x eigenfunction and eigenvalue. It can be proved that for hermitian operators eigenvalues are aalways real ( tutorial problem). Eigenfunction and eigenvalues plays a very important role for a a quantum mechanical system. Let us first give the mathematical statement then we will go into the physical meaning.

, If a wavefunction ( x) is not an eigenfunction of some operator A

associated with some physical observable, it can be expanded as = c n n , where n s are the eigenfunctions of that operator with

A n = a n n . Then the
the probability of occurrence of a n , P (a n ) =| c n | 2 . If n | m = mn ,

| a
n

|2 = 1 .

Let us try to physically interpret the above statement. Suppose we are prepare an electron or proton or something in a quantum state and make an measurement for a given observable ( momentum, polarization or position). If that state is an eigenstate of the operator that is associated with that observable, we will always get the same result. This is the significance of being an eigenstate. On the otherhand if the above mentioned state is not an eigenstate, everytime we shall make an measurement of that observable the system will collapse in one of the eigenstates of the operator, that is associated with that particular observable. In which eigenstate it will collapse , cannot be determined with absolute certainty, but can be given a probability density which is proportional to the absolute square of expansion coefficient of that state in terms of the eigenstates of the corresponding operator. The number of eigestates can be either finite or infinite. Now with this interpretation in hand we can easily explain the results of the electron interference experiment as well as the photon polarization

experiment. In the former , the number of such eigenstates of the postion operator is infinite. The electron after coming out of either of the slits can fall anywhere on the screen with certain probability. Any position on the screen ( x co ordinate ) is an eigenstate of the position operator. Thus the intensity of the interference pattern which is a function of the position is here equivalent to | a n | 2 . Only here the index n form a continuum. In the case of the photon polarization problem, the number of the eigenstates are two associated with the two mutually perpendicular direction of polarization. Therefore even though the photon is initially not in an eigenstate of the polarization operator, after the measurement it collapse in one of eigenstates with the corresponding probability. Here the index n is discrete and can take only two values. Thus we can make the following statement: , iff the For a given observable, associated with an hermitian operator A state is an eigenfunction , with eigenvalue a , then the state has unique value for the observable A ( see we have used a slightly different notation to distinguish between the observable and its corresponding operator), then measurement of A on identical samples with same , will always give the same value a . Mathematically

A | = a | example : ih | e ikx = hk | e ikx x Let us now learn certain mathematical properties associated with these Hermitian operators, which will play an important role in the rest of our discussion.
| = | A | = A | Definition | A
Property 1 : Eigenvalues are real ( that should be, since eigenvalues are values which we shall obtain after making measurement of system for different physical quantities or observable like momentum, energy etc. Thus they have to be real. Since hermitian operators or matrices satisfy this property, only they can be associated with the observables. Proof: Show yourself (tutorial problem).

Property 2. Non degenerate eigen-functions ( having different eigenvalues) are mutually orthogonal. | = a | ; A n n n | =a | A
m m m

| = a | n | A m m n m

Using the definition of hermitian operator we obtain An | m = a n n | m = a n n | m Since these two are equal because of hermiticity, subtracting each other we get (a m a n ) n | m = 0 Since a m a n , n | m = 0 Property 3: A complete ortho-normal set can be formed using eigenfunctions . for any Hermitian operator A The procedure is to choose any set of basis states { n } . Then we need to calculate the matrix element of the hermitian operator in that basis, namely | = A . Then we need to diagonalize the matrix to find out the n | A m mn eigenvectors. They will form an orthonormal basis. A well known example is based on our knowledge Fourier transform. Consider the momentum operator: = ih , and consider functions = 1 e ikx . A k x 2 They satisfy the eigenvalue equation since ih | k = hk | k x It can be easily checked that they are orthonormal, since

k | k' = (k k' ) .
( x) =
1

Any function can be expanded as or | = a(k ) | k 2 k Thus all the momentum eigenstates form an orthonormal basis.

dka(k )e

ikx

Property 4. , the expand it in the orthoIf is not an eigenstate of the operator A . = c , where normal set using the eigenstates of the operator A n n
n

A | n = a n | n and c n = n | .
The probability

P (a n ) =| c n | 2
When we get a value a n jump from to n . Property 5: Resolution of identity | = c n | n >= | n n | 1 = | n n |
n n n

, the measurement process makes the state

Thus the sum of outer products of all the eigenstates with itself is same as identity. This is known as resolution of identity. If the index of eigenstate is , then the sum is continuous ( for example eigenstates of position operator x replaced by an integration, dx | x x |= 1 Property 6: Commuting observables. Two observables are called commuting observables, if the associated B |B | . &B | = A = B A satisfy for all | , A operators A The commutator of two operators are defined as B = [A , B B A ] . A , p x ] = [ x , i h ] = i h For example the operators [ x x Position and momentum operator do not commute with each other. Acknowledgement: These notes are prepared mostly from the lecture notes of Prof. R. Rajaraman and the Principles of Quantum Mechanics by P. A. M. Dirac.