Portland State University

PDXScholar

Physics Faculty Publications and Presentations Physics

6-1998

Path Integral for the Quantum Harmonic Oscillator

Using Elementary Methods
Scott M. Cohen
Portland State University

Follow this and additional works at: https://pdxscholar.library.pdx.edu/phy_fac

Part of the Physics Commons

Let us know how access to this document benefits you.

Citation Details
Cohen, S. M., (1998). Path integral for the quantum harmonic oscillator using elementary methods.
American Journal of Physics, 66, 537-540

This Article is brought to you for free and open access. It has been accepted for inclusion in Physics Faculty
Publications and Presentations by an authorized administrator of PDXScholar. Please contact us if we can make
this document more accessible: [email protected].
Path integral for the quantum harmonic oscillator using
elementary methods
S. M. Cohen
Department of Physics, Portland State University, Portland, Oregon 97207
~Received 12 September 1997; accepted 12 November 1997!
We present a purely analytical method to calculate the propagator for the quantum harmonic
oscillator using Feynman’s path integral. Though the details of the calculation are involved, the
general approach uses only matrix diagonalization and well-known integrals, techniques which an
advanced undergraduate should understand. The full propagator, including both the prefactor and
the classical action, is obtained from a single calculation which involves the exact diagonalization
of the discretized action for the system. © 1998 American Association of Physics Teachers.

I. INTRODUCTION ing bond lengths confined in a harmonic potential. Addition-

ally, we show how the classical action arises naturally, along
Since their introduction,1 Feynman path integrals have be- with the prefactor, from a single calculation. This differs
come a powerful method of calculation for quantum me- from previous approaches in which only the prefactor was
chanical problems.2,3 Though until recently exact solutions calculated, the appearance of the classical action being as-
were available for only the simplest cases, great advances in sumed due to a theorem given by Feynman.2
developing methods of solving these integrals have been
made in the last 15 years.4 Yet even before these advances,
the approach bore fruit in many ways. For example, the
II. FORMAL EVALUATION OF THE PATH
derivation5 of the ‘‘Feynman rules’’ was an extremely im-
portant contribution which greatly simplified calculations in INTEGRAL
perturbation theory.
The quantum propagator, K(b,a) for a particle beginning
In a recent article, English and Winters6 have presented a
method of calculating the Feynman path integral for the pref- at position x(t a )5a and ending at x(t b )5b, is given as2
actor of the propagator of the quantum harmonic oscillator.
The motivation for their work was ‘‘to introduce a formula-
tion of quantum mechanics which is usually considered be-
K ~ b,a ! 5 E D @ x ~ t !# exp S E
i
\
tb

ta
D
L @ x ~ t ! ,ẋ ~ t !# , ~1!

yond the scope of most undergraduate courses.’’ We agree where

with these authors that it is of interest to make alternative
approaches to quantum mechanics accessible to the under- L @ x ~ t ! ,ẋ ~ t !# 5 21 mẋ 2 2 21 m v 2 x 2 ~2!
graduate. We believe that path integrals have great beauty in
the simplicity of their basic formulation. They also clarify is the classical Lagrangian, and the symbol * D @ x(t) # repre-
various aspects of quantum mechanics, such as the uncer- sents integration over all paths in configuration space begin-
tainty principle. The clarification in this particular case fol- ning at a and ending at b. As is common practice, these
lows immediately from the central idea upon which the path integrals may be done by first partitioning the time interval
integral formulation is based: that all paths in configuration into N pieces of width e each, so that T5t b 2t a 5N e . At the
space contribute to the evolution of the wave function. Thus end of the calculation the limits N→`, e →0, are taken,
there is an intrinsic uncertainty as to the evolution of any such that T5N e is held constant. Then, with x j 5x( j e ), we
system ~we cannot know the trajectory the system follows!, may write

S D
and this uncertainty is explicitly illustrated in this approach. N/2
In this note, we give an alternative presentation which we m
K ~ b,a ! 5 lim
believe is somewhat more direct than that of English and N→` 2 p i\ e
Winters. The method used by these authors required the use e →0
of a symbolic computational program, and an intermediate
result written in terms of continued fractions. ~But see our
Appendix for a discussion of how the approach of these au-
3 E E
`

2`
•••
`

2`
dx 1 dx 2 •••dx N21

thors may be completed analytically.! Our method does not N 22e2v2x2

require the use of a computer and is straightforward, for- 3e im/2\ e ( j51 $ ~ x j 2x j21 ! j% . ~3!
mally, so it should be accessible to students. An understand- The argument of the exponential contains the quadratic form
ing of Gaussian integrals, and of matrices and their eigen-
N
vectors and eigenvalues, are the only prerequisites to
following this approach. Q5 (
j51
@~ x j 2x j21 ! 2 2 e 2 v 2 x 2j #
Although this problem has been addressed in numerous
other works,2,3,6,7 our presentation is new in some important 5x 20 1x 2N 2 e 2 v 2 x 2N 22x 1 x 0 22x N x N21 1Q 8 , ~4!
ways. First, we discretize the action from the very beginning,
allowing us to obtain a final result which is exact for arbi- where we may write, Q 8 5xW T AxW . Here,
trary N ~the number of intervals chosen for the xW T 5 ~ x 1 x 2 •••x N21 ! ~5!
discretization—see below!. These results are thus directly
transferable to the case of a polymer chain with nonvanish- is the transpose of xW , and

537 Am. J. Phys. 66 ~6!, June 1998 © 1998 American Association of Physics Teachers 537
 S D S D H F
22 e 2 v 2 21 0 0 ••• m N/2
im
K ~ b,a ! 5 lim exp x 2 1x 2 2 v 2 e 2 x 2N
21 22 e v 2 2
21 0 ••• N→` 2 p i\ e 2\ e 0 N
e →0
21 22 e 2 v 2 21 •••

GJ
A5 0 .
N21
21 22 e 2 v 2 ~ ON21,j x N 1O1,j x 0 ! 2
0 0 2 ( lj
A A A  j51

FE S D G
~6! N21
` im
3 )
j51 2`
exp l z 2 dz j .
2\ e j j
~14!
If, using a change of variables from the x j to new variables
z j , we can rewrite Q 8 into the form The integrals are now simple Gaussians, as advertised above,
yielding

S D H F
N21

(
1/2
Q 8 5constant1 l j z 2j , ~7! m im
K ~ b,a ! 5 lim exp x 2 1x 2 2 v 2 e 2 x 2N
j51
N→` 2 p i\ e 2\ e 0 N
e →0

GJ S ) D
where the constant does not depend on the z j , then the N21 N21 21/2
~ ON21,j x N 1O1,j x 0 ! 2
~coupled! integrals in Eq. ~3! will have been reduced to N
21 separate Gaussian integrals. First, we will find a trans-
2 (
j51 lj j51
lj

S D
formation of variables, 1/2
m
5e i/\ S cl lim ~ det A! 21/2[F ~ T ! e i/\ S cl.
xW 5OyW , ~8! N→` 2 p i\ e
e →0

such that ~15!

We will show below that
OT AO5L,

with L a diagonal matrix, L i j 5l j d i j , and O will be or-

~9!
S cl5 lim
N→`
e →0
m
2e 0 NF
x 2 1x 2 2 v 2 e 2 x 2N

G
thogonal since A is symmetric and real. Then we may write
N21
~ ON21,j x N 1O1,j x 0 ! 2
N21 2 ( lj
~16!
(
j51
Q5x 20 1x 2N 2 v 2 e 2 x 2N 22x N ON21,j y j
j51 is indeed the classical action, as it must be; and we will find
N21 N21 det A and thus the prefactor F(T), as well.
22x 0 (
j51
O1,j y j 1 (
j51
l j y 2j . ~10!

III. DIAGONALIZATION OF THE MATRIX, A

Completing the squares, we change variables once again to
To implement the transformation of variables, Eq. ~8!, we
x N ON21,j 1x 0 O1,j must find the matrix O which diagonalizes A. As is well
z j 5y j 2 , ~11! known, O is the matrix of the eigenvectors of A. It is not
lj
difficult to show that a complete set of eigenvectors, which
we shall denote as eW j , is given in terms of their components
yielding
by

Q5
N21

(
j51
l j z 2j 1x 20 1x 2N 2 v 2 e 2 x 2N ~ eW j ! i 5 A S D
2
N
sin
pi j
N
, ~17!

N21 with 1<i, j<N21. The corresponding eigenvalues are

~ ON21,j x N 1O1,j x 0 ! 2
2 (
j51 lj
. ~12!
l j 522 v 2 e 2 22 cos S D pj
N
54 sin2
pj
2N S D
2 v 2e 2, ~18!

Since O is orthogonal, det O51, and the Jacobian of both leading directly to the result

FS S D DG
transformations, Eqs. ~8! and ~11!, is unity. Hence we have N21 N21
pj
the replacement det A5 ) l j 5 j51
j51
) 4 sin2
2N
2 v 2e 2

E E
`

2`
•••
`

2`
dx 1 dx 2 •••dx N21 ⇒ E E
2`
`
•••
`

2`
dz 1 dz 2 •••dz N21 ~ 24 ! N21
5 2 )F
M /2

sin2 u 2sin2 S DG
pj
, ~19!
sin u 21 j51 M
~13!
where we have written sin u for v e /25 v T/2N, and M
in Eq. ~3! along with the transformation of variables. We 52N. The product appearing in the final form of this equa-
obtain tion is given in Hansen,8

538 Am. J. Phys., Vol. 66, No. 6, June 1998 S. M. Cohen 538
 F S DG
M /2 The evaluation of a 0 is identical to that of a N , apart from the
pj
)
j51
sin2 u 2sin2
M
5 ~ 21 ! M /22 12M sin~ M u ! cot u , term proportional to 1/N; that is,

S D
~20!
mN cos~~ N22 ! f ! 2cos~ N f !
and we find 9 a 05 12 . ~28!
2T 2 sin f sin~ N f !
sin~ 2N u !
det A5 , ~21! In the same way, we find
sin~ 2 u !
N21
correct for all N> v T/2. Furthermore, taking mN ON21,j O1,j
Oi j 5 ~ eW j ! i , ~22!
a 0N 52
T (
j51 lj

we may calculate pj
~ 21 ! j sin2 S D
F
N21
m N mN sin f
(
S D
mN 2 2 v 2 T 2 2 5 52 .
S cl~ N ! 5 x 0 1x N 2 2 x N T j51 pj T sin~ N f !
2T N cos f 2cos

G
N
N21
~ ON21,j x N 1O1,j x 0 ! 2 ~29!
2 (
j51 lj
In writing the last equality, we have again referred to
5a N x 2N 1a 0 x 20 1a 0N x 0 x N . ~23! Hansen,11

S D
Now,
pj

a N5
mN
2T S v 2T 2
12 2 2
N j51 lj
N21
O2N21,j
( . D ~24!
N21

(
j51
~ 21 ! j sin2

cos f 2cos
N
pj
S D
52N sin f csc N f . ~30!

But, N

N21
O2N21,j 2
N21 S
p j ~ N21 !
sin2
N D IV. EXACT PROPAGATOR FOR N DISCRETE
( (
S D
5 TIME INTERVALS
j51 lj N j51 p j
4 sin2 2sin2 u
2N Using the above results, we can determine the propagator

S D
for an arbitrary number, N, of divisions of the time interval
pj T. This expression may be useful for students and others
N21 sin2
2 N
(
doing numerical work with path integrals, as a check of their

S D
5
N j51 pj discrete-time algorithms. The result is
4 sin2 2sin2 u

S D
2N
mN sin~ 2 u ! 1/2

1
M /221 12cos S D
4p j
M
K N ~ b,a ! 5
2 p i\T sin~ 2N u !
e i/\ S cl~ N ! , ~31!

(
S D
5 , ~25! with
2N j51 2p j
cos f 2cos

HS D
M
mN cos~~ N22 ! f ! 2cos~ N f ! 2
with, as before, M 52N, sin u5vT/2N, and also cos f51 S cl~ N ! 5 12 a
2T 2 sin f sin~ N f !

S D
2 v 2 T 2 /2N 2 . This sum may also be found in Hansen,10
v 2 T 2 cos~~ N22 ! f ! 2cos~ N f ! 2
M /221
2 p jk
cos S D 1 12
N2
2
2 sin f sin~ N f !
b

J
M
(
S D
2 sin f
j51 2p j 2 ab , ~32!
cos f 2cos sin~ N f !
M

52
M
2
csc f csc
Mf
2
cos HS D JM
2
2k f u 5arcsin S D vT
2N
, ~33!

1
1
4
csc2
f
2
1
SD
2 ~ 21 ! k sec2
4
f
2
, SD ~26! and

where in our case we need k50,2. Then we have f 5arccos 12

v 2T 2
2N 2 S. D ~34!

a N5
mN
2T
12 2 2
N S
v 2 T 2 cos~~ N22 ! f ! 2cos~ N f !
2 sin f sin~ N f !
. D Note that we have used x 0 5a, x N 5b.
~27! Finally, the true propagator is obtained as

539 Am. J. Phys., Vol. 66, No. 6, June 1998 S. M. Cohen 539
 K ~ b,a ! 5 lim K N ~ b,a !

5
N→`

S mv
2 p i\ sin v T D 1/2
N21

)
j51
Aj
5
B j j51
N21

) H sin@ j arccos~ g /2 !#
sin@ arccos~ g /2 !#
sin@~ j11 ! arccos~ g /2 !#
sin@ arccos~ g /2 !#
J
3exp H im v
2\ sin v T
@~ a 2 1b 2 ! cos v T22ab # , J 5
sin@ arccos~ g /2 !#
sin@ N arccos~ g /2 !#
. ~A5!

~35! Thus we have

H J
having used the fact that for large N, F~ T !
vT
F S
v 2T 2
DG 1/2

S D
u> , 1/2 sin arccos 12
2N m 2N 2

F S DG
5 lim N
and 2 p i\T N→` v 2T 2
sin N arccos 12

H J
vT 2N 2

F G
f> .
N vT 1/2

S D 1/2 sin
In conclusion, we have presented a strictly analytical method m N

F S DG
5 lim N
by which the full propagator for the quantum harmonic os- 2 p i\T N→` vT
cillator may be obtained using Feynman’s path integral ap- sin N
N
proach. Though the details are involved, the general ap-
proach should be accessible to advanced students. In
particular, our presentation may be of interest to those in-
structors of graduate-level quantum mechanics who would
5 S m
2 p i\T DH 1/2
vT
sin~ v T ! J 1/2
. ~A6!

like to introduce path integrals into their courses. Finally, then

ACKNOWLEDGMENTS F~ T !5 S mv
2 p i\ sin~ v T ! D 1/2
. ~A7!
We would like to thank Pui-Tak Leung for helpful com-
This is the desired result, once again obtained by purely ana-
ments and for bringing Ref. 6 to our attention.
lytical means.
APPENDIX
1
R. P. Feynman, ‘‘Space-time approach to nonrelativistic quantum mechan-
In Ref. 6, it is shown that the prefactor, F(T), may be ics,’’ Rev. Mod. Phys. 20, 367–387 ~1948!.
written as a product of factors, 2
R. P. Feynman and A. R. Hibbs, Quantum Mechanics and Path Integrals

S D F G
1/2 N21 1/2 ~McGraw–Hill, New York, 1965!.
m Aj
) L. S. Schulman, Techniques and Applications of Path Integration ~Wiley-
3
F~ T !5 lim N , ~A1!
2 p i\T N→` j51 Bj 4
International, New York, 1981!.
C. Grosche and F. Steiner, ‘‘How to solve path integrals in quantum me-
where the A j and B j satisfy the same recursion relations, chanics,’’ J. Math. Phys. 36, 2354–2385 ~1995!, and references therein.
5
R. P. Feynman, ‘‘An operator calculus having applications in quantum
A j 5 g A j21 2A j22 , ~A2! electrodynamics,’’ Phys. Rev. 84, 108–128 ~1951!.
6
L. Q. English and R. R. Winters, ‘‘Continued fractions and the harmonic
B j 5 g B j21 2B j22 , ~A3! oscillator using Feynman’s path integrals,’’ Am. J. Phys. 65, 390–393
~1997!.
with g 522 v 2 T 2 /N 2 , and starting conditions, A 21 521, 7
J. T. Marshall and J. L. Pell, ‘‘Path-integral evaluation of the space-time
A 0 50, B 21 50, and B 0 51. As these authors observe, A j11 propagator for quadratic Hamiltonian systems,’’ J. Math. Phys. 20, 1297–
5B j . What we would like to point out is that these relations 1302 ~1979!.
8
Eldon R. Hansen, A Table of Series and Products ~Prentice-Hall, Engle-
brand these objects as Chebyshev polynomials of the second wood Cliffs, NJ, 1975!, p. 497, Eq. #91.1.18.
kind.12 Specifically, 9
This result has been obtained previously in H. Kleinert, Path Integrals in

B j 5U j SD g
2
5
sin@~ j11 ! arccos~ g /2 !#
sin@ arccos~ g /2 !#
. ~A4! 10
11
Quantum Mechanics, Statistics, and Polymer Physics ~World Scientific,
Singapore, 1990!, p. 85, Eq. #2.140.
Reference 8, p. 272, Eq. #41.2.19.
Reference 8, p. 272, Eq. #41.2.25.
Therefore, the product appearing in the formula for F(T) 12
I. S. Gradshteyn and I. M. Rhyzik, Table of Integrals, Series, and Products
is just ~Academic, Orlando, FL, 1980!, pp. 1032–1033.

540 Am. J. Phys., Vol. 66, No. 6, June 1998 S. M. Cohen 540