You are on page 1of 8

Gaussian beams in the optics course

Enrique J. Galvez

Citation: Am. J. Phys. 74, 355 (2006); doi: 10.1119/1.2178849


View online: http://dx.doi.org/10.1119/1.2178849
View Table of Contents: http://ajp.aapt.org/resource/1/AJPIAS/v74/i4
Published by the American Association of Physics Teachers

Related Articles
Modeling the Dynamics of Gel Electrophorresis in the High School Classroom
Phys. Teach. 51, 28 (2013)
Conductance quantization: A laboratory experiment in a senior-level nanoscale science and technology course
Am. J. Phys. 81, 14 (2013)
Reflection of a polarized light cone
Am. J. Phys. 81, 24 (2013)
Using iPads to illustrate the impulse-momentum relationship
Phys. Teach. 51, 54 (2013)
The Fuse-Wires-in-Parallel Problem
Phys. Teach. 51, 38 (2013)

Additional information on Am. J. Phys.


Journal Homepage: http://ajp.aapt.org/
Journal Information: http://ajp.aapt.org/about/about_the_journal
Top downloads: http://ajp.aapt.org/most_downloaded
Information for Authors: http://ajp.dickinson.edu/Contributors/contGenInfo.html

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
APPARATUS AND DEMONSTRATION NOTES
Jeffrey S. Dunham, Editor
Department of Physics, Middlebury College, Middlebury, Vermont 05753

This department welcomes brief communications reporting new demonstrations, laboratory equip-
ment, techniques, or materials of interest to teachers of physics. Notes on new applications of older
apparatus, measurements supplementing data supplied by manufacturers, information which, while not
new, is not generally known, procurement information, and news about apparatus under development
may be suitable for publication in this section. Neither the American Journal of Physics nor the Editors
assume responsibility for the correctness of the information presented. Submit materials to Jeffrey S.
Dunham, Editor.

Gaussian beams in the optics course


Enrique J. Galvez
Department of Physics and Astronomy, Colgate University, Hamilton, New York 13346
共Received 19 September 2005; accepted 27 January 2006兲
We introduce the physics of high-order Gaussian beams to the treatment of Gaussian beams in the
undergraduate optics course. Of particular interest are Laguerre-Gauss beams, which provide the
basis for discussing the new and increasingly important concept of the orbital angular momentum of
light. We also describe laboratory exercises that complement the class material. © 2006 American
Association of Physics Teachers.
关DOI: 10.1119/1.2178849兴

I. INTRODUCTION interesting families of Gaussian beams are the Laguerre-


Gauss beams, which are the solutions of the wave equation
Optics is a popular choice as an upper-level elective in cylindrical coordinates. They bring a new fundamental
course in the undergraduate physics curriculum. It is a topic concept to optics: orbital angular momentum. This is the
with many interesting phenomena and applications. The so- angular momentum of the light that arises when the beam
lutions of problems are often elegant and at a mathematical has a helical wave front.1 These types of beams also serve as
level that undergraduates can easily grasp. Since optics is the an introduction to an interesting new field in optics: singular
basis of many scientific and technological applications, stu- optics.2,3
dents leave the course with a solid understanding of optics In this article we present the framework for covering
fundamentals and with a useful set of laboratory skills. Gaussian beams to include the discussion of high-order
The standard optics course contains several core topics: beams and orbital angular momentum. This treatment reflects
light waves, geometric optics, polarization, interference, and our own efforts and experience at incorporating high-order
diffraction. It is up to the instructor to decide what other Gaussian beams into our optics course at Colgate University.
topics to include 共e.g., Fourier optics, holography, coherence, In Sec. II we describe how to introduce the fundamental
lasers, nonlinear optics, and quantum optics, to name a few兲. zero-order solution of the paraxial wave equation. We con-
However, optics has evolved tremendously in the last few tinue in Sec. III with a discussion of the high-order solutions
decades. New subfields and technologies have appeared in of the wave equation in rectangular coordinates: Hermite-
response to major discoveries. Therefore, it is important to Gauss beams. The solution of the wave equation in cylindri-
reevaluate the evolution of the core material and the selec- cal coordinates, giving rise to Laguerre-Gauss beams, is pre-
tion of specialty topics. sented in Sec. IV. Section V presents a discussion of the new
The advent of the laser has had a huge impact in optics. It form of angular momentum carried by Laguerre-Gauss
has led to technological applications in most fields of science beams. In all sections we present laboratory experiments that
and engineering. The teaching laboratories of the optics can be used to complement the class material.
course have been significantly enhanced with the use of la-
sers. Yet laser fundamentals and laser-beam propagation do
not receive prominent roles in most optics textbooks. The
topic of Gaussian beams traditionally describes the charac-
teristics of laser beams and their propagation. It is important II. ZERO-ORDER GAUSSIAN SOLUTION
for finding useful parameters of the light beam and its propa-
gation through an optical system. Optics courses often start with a review of waves and the
In the last decade we have seen the emergence of an ex- wave equation 共see, for example, Ref. 4兲. This begins with a
tension of Gaussian beams: high-order Gaussian modes. discussion of light waves propagating in one dimension. The
These are high-order solutions of the paraxial wave equation. wave has a phase given by 共kz − ␻t兲 when propagating along
High-order Gaussian modes of light are a hidden treasure. the z direction, where k = 2␲ / ␭ is the wave number, ␻ is the
They are rich with physics and wave phenomena. The most angular frequency, and ␭ is the wavelength. The discussion

355 Am. J. Phys. 74 共4兲, April 2006 http://aapt.org/ajp © 2006 American Association of Physics Teachers 355

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
evolves into a treatment of waves propagating in three di-
mensions. Two types of three-dimensional waves are particu-
larly useful: plane waves and spherical waves. A plane wave
w共z兲 = w0 冑 1+
z2
zR2
, 共5兲

is in essence a one-dimensional wave propagating in an ar-


where w0 is the waist of the beam and zR = ␲w20 / ␭ is the
bitrary direction, with a phase given by 共k · r − ␻t兲, where k is
Rayleigh range. This solution agrees with the terms obtained
the propagation vector and r is the position. A spherical wave intuitively before. It contains an amplitude that decreases
propagates symmetrically from the origin. It has a phase axially with z via w共z兲, and transversely it falls as a function
given by 共kr − ␻t兲 and an amplitude proportional to 1 / r,
of x and y via the Gaussian term.
where r = 兩r兩. It is important to follow this solution with a discussion of
A laser beam is a narrow collimated beam of light that Gaussian beam propagation and the effect of lenses on
diverges slowly as it propagates. The wave function that de- Gaussian beams of various widths and wave-front radii R共z兲.
scribes it is a solution of the paraxial wave equation. We
This formalism predicts the correct size and location of the
found it useful to begin the discussion of Gaussian beams by
beam waist.5,6 The discussion of Gaussian-beam propagation
first estimating the main terms in the solution.5 This is a
can be either short, using Ref. 5 as a basis, or expanded by
healthy exercise because it invokes intuition and helps stu-
introducing the elegant “ABCD” matrix formalism.7
dents understand the exact form of the solution. Gaussian
The treatment of Gaussian beams also brings to the fore-
beams must be represented by a wave function that propa-
front a discrepancy between the predictions of wave optics
gates mainly in one dimension and very slowly in the other
and those of ray optics: the finite waist w0 of the beam.
two dimensions. Thus, the wave function has the character-
Geometric optics, which uses ray optics for describing imag-
istics of both plane and spherical waves. The two important
ing with lenses, predicts a zero waist at the focal point of a
components of a wave function are its amplitude and phase.
lens when a collimated beam is incident on the lens. It is
Let us assume that the main axis of the light beam is aligned
wave optics that gives the correct beam width at the focal
along the z direction. Because of the small but finite expan-
point. Ignoring a discussion of Gaussian beams leads to mis-
sion of the beam, its amplitude must decrease with z. Since
conceptions, and can give the wrong results in the
the amplitude is restricted to small values of x and y, then it
laboratory.6 Usually the instructor can remedy this deficiency
must fall rapidly as x or y increase. The Gaussian shape
when discussing diffraction and resolution, but this seems
specified by exp关−共x2 + y 2兲 / w2兴, where w is the width of the
unnecessarily qualitative when the exact solution is at hand.
beam, seems intuitively appropriate. Because of the mostly In addition, the discussion on the waist of the Gaussian beam
one-dimensional propagation the main phase term should be can be used to reinforce the concept of resolution in the
of the form 共kz − ␻t兲 for points along the axis, but because of chapter on diffraction.
the expansion of the beam, its wave front, the surface con- A useful laboratory exercise on this subject is to challenge
taining points of equal phase, it must be spherical. Therefore, students to send a laser beam through a small pinhole. We
the phase must also contain a term that resembles a spherical give the students a laser, an assortment of lenses, a pinhole
wave for finite values of x and y. 共e.g., 50 ␮m in diameter兲, and a detector. The exercise is
After this step one then proceeds with a more formal conceived in such a way that in order to complete the experi-
method of solving the wave equation. The solution is set up ment students must expand the beam. Successful execution
to have the form of the experiment also necessitates a measurement of the
divergence of the laser beam; this could be an exercise that is
⌿共x,y,x,t兲 = E共x,y,z兲ei共kz−␻t兲 . 共1兲 assigned before the challenge.
The phase given by k共x2 + y 2兲 / 关2R共z兲兴 in Eq. 共4兲 accounts
The steps that follow consist of deriving the Helmholtz equa- for the spherical wave front of the beam. The radius of cur-
tion and making the approximations applicable to Gaussian
vature of the wave front, given by R共z兲 = z + zR2 / z, has a de-
beams. The derivation leads to the paraxial wave equation,
pendence on z that is interesting in itself. The Gouy phase,
⳵ 2E ⳵ 2E ⳵E given by ␾共z兲 = tan−1共z / zR兲, is also interesting. It is a conse-
+ + 2ik = 0. 共2兲 quence of the propagation of the Gaussian beam, and
⳵x 2
⳵ y2 ⳵z changes most rapidly when the beam goes through a focal
point.
To obtain the zero-order solution of Eq. 共2兲 we first introduce The phase of the wave is measured by interference. Thus,
a trial solution of the form the properties of Gaussian beams can be revisited later in the
2+y 2兲/关2q共z兲兴 ip共z兲 course when covering the chapter on interference.
E0共x,y,z兲 = Aeik共x e . 共3兲 A convenient optical setup to study wave-fronts is a Mach-
Zehnder interferometer,8 shown in Fig. 1. The setup and
The solution of Eq. 共3兲 then reduces to finding q共z兲 and p共z兲. alignment of the Mach-Zehnder interferometer is a lab exer-
The trial solution of Eq. 共3兲 fixes the dependence on the cise in itself. This alignment job can be quite straight for-
transverse coordinates x and y, forcing the solution to have a ward if done systematically with an iris aligned with the
symmetric transverse profile. A complex solution for q will holes of an optical breadboard. This method is described in
lead to the terms introduced intuitively above. The final so- detail in the Appendix. We find nonpolarizing cube beam
lution is5 splitters convenient for use in interferometers. They are rea-
sonable in cost compared to other optical components 共e.g., a
A −共x2+y2兲/w共z兲2 ik共x2+y2兲/关2R共z兲兴 i␾共z兲 beam splitter of 1 cm in size, such as Melles Griot part num-
E0共x,y,z兲 = e e e . 共4兲
w共z兲 ber 03BSC003, costs about $150兲. It is also good practice to
use laser beams polarized along the vertical or horizontal
The width of the beam or spot size is given by directions. If the input polarization is oriented at an arbitrary

356 Am. J. Phys., Vol. 74, No. 4, April 2006 Apparatus and Demonstration Notes 356

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
interference pattern produced by the focused beam and the
unfocused beam coming from the other arm of the interfer-
ometer. By moving the camera along the propagation axis of
the beams about the beam waist one can see the fringes shift-
ing due to the Gouy phase. Note that this is done without
changing the path-length difference of the light in the inter-
ferometer.
We can make the interferometer very stable against vibra-
Fig. 1. Mach-Zehnder interferometer for studying the phases of optical tions by using “pedestal” mounts, and by making the dimen-
beams. The basic components are two beam splitters 共BS兲 and two mirrors sions of the interferometer as small as possible. By mounting
共M兲, with one mirror mounted on a translation stage 共T兲. A lens 共L兲 can be one of the mirrors on a translation stage and putting a piezo-
added to one of the arms for observing the pattern produced with the light electric ceramic as a spacer in the stage, one can scan the
from the other arm. A computer-generated grating 共G兲 can be inserted in one
of the arms to produce high-order modes.
interference fringes by applying a voltage to the piezoelectric
共e.g., the piezoelectric from Thorlabs model AE0505D8 costs
about $130兲. The 0-to-150-volt driver of the piezoelectric is
a low-cost high-voltage amplifier 共e.g., EMCO High Voltage
angle, differences in the reflection coefficients of the beam
part number Q02-24 costs about $80兲. An inexpensive way
splitters and mirrors may produce changes in the state of
to scan the fringes is to attach a rubber band to the back of
polarization that depend on the interferometer path taken by
one of the mirror mounts and then to pull the other end of the
the light, and thus reducing the visibility of the interference
rubber band in a systematic way.
pattern.
When the beams coming from both arms of the interfer-
ometer are not exactly collinear the interference pattern con- III. HIGH-ORDER HERMITE-GAUSS SOLUTIONS
sists of straight fringes, as shown in Fig. 2共a兲. These can
either be projected onto a screen with a diverging lens placed The high-order solutions of the paraxial wave equation
after the interferometer, or sent to a CCD camera without a depend on the system of coordinates that we use. It is natural
lens. By inserting a diverging lens in one of the arms we can to start the discussion with the solution in rectangular coor-
investigate the spherical wave front of the beam: the inter- dinates. Rectangular solutions of the wave equation also
ference of collinear beams with different radii of curvature have a familiar mathematical formalism for the students.
produces the “bulls-eye” pattern shown in Fig. 2共b兲. We start by revisiting the paraxial wave equation 关Eq. 共2兲兴,
One can also use the interferometer to observe the Gouy and introducing the more general trial solution:
phase. This is done by placing a converging lens in one of 2+y 2兲/关2q共z兲兴 ip共z兲
the arms of the interferometer such that it focuses the beam E共x,y,z兲 = Ag共x,z兲h共y,z兲eik共x e . 共6兲
going through that arm outside the inteferometer. A CCD
Note that now we allow the solution to have independent x
camera placed in the path of the beams will show a bulls-eye
and y dependencies. Substituting Eq. 共6兲 into Eq. 共2兲
yields5,7,9 two Hermite differential equations for g and h. The
final solution is
Em,n共x,y,z兲

=
A
w共z兲
Hm 冉 冊冉 冊
冑2x
w共z兲
Hn
冑2y
w共z兲
2+y 2兲/w共z兲2 ik共x2+y 2兲/关2R共z兲兴 i␾共z兲
⫻e−共x e e . 共7兲
A beam described by this solution is known as a Hermite-
Gauss beam. The indices m and n of the Hermite polynomi-
als Hm and Hn label particular solutions or modes. We define
the order of the solution by N = n + m. Modes of the same
order are degenerate in laser resonators. Since H0 = 1, the
solution of Eq. 共4兲 for n = m = 0 is the zero-order solution
described in the preceding section. Aside from the Hermite
polynomials, the other terms of Eq. 共7兲 are the same as those
of Eq. 共4兲 except for the Gouy phase, which is given by a
modified expression, ␾共z兲 = 共N + 1兲tan−1共z / zR兲. The depen-
dence of the Gouy phase on the order N gives rise to the
higher resonant frequencies of 共high-order兲 transverse modes
in laser resonators. The constant term that normalizes the
total intensity is given by10 A = 关21−N / 共␲n ! m ! 兲兴1/2.
The Hermite-Gauss solutions have a variety of interesting
Fig. 2. Pictures of images of interference patterns produced with the Mach- shapes that can be investigated in the laboratory by use of an
Zehnder interferometer and recorded with a CCD camera. In patterns 共a,b兲,
共c,d兲, and 共e,f兲 the signal beams were zero-order Gaussian, first-order
open-frame laser with a thin wire placed in the cavity.10,11
Hermite-Gauss, and first-order Laguerre-Gauss, respectively. In patterns Losses by scattering off the wire cause the laser to oscillate
共a,c,e兲 the reference beam was unexpanded and noncollinear, and in 共b,d,f兲 in a mode that has its nodes along the wire. An alternative
the reference beam was expanded and collinear. method is to use a thin coverslip as an etalon and frequency

357 Am. J. Phys., Vol. 74, No. 4, April 2006 Apparatus and Demonstration Notes 357

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
Fig. 3. Computer-generated holograms to generate first-order Hermite-
Gauss beams 共a兲, and Laguerre-Gauss beams 共b兲. The lower images are
photos of the diffraction patterns that result when the gratings above are
illuminated by a zero-order beam. The gratings used to produce the patterns Fig. 4. Perspective view of the wave front of a first-order Laguerre-Gauss
in the images did not have the same density of fringes. beam. The helical surface is formed by points where the wave has the same
phase 共crests, in the picture兲.

select the mode by tilting the etalon.12 By sending the beam


through an interferometer one can examine the ␲ phase dif-
ference between adjacent lobes, as described below.
A convenient way to produce high-order modes is to use p and ᐉ label the solutions of order N = 2p + 兩 ᐉ 兩, which also
computer-generated holograms. Binary amplitude holograms reduce to the zero-order solution when p = ᐉ = 0. The normal-
are simple to make, but require a technology that is increas- ization constant is given by10 A = p ! 兵2 / 关␲ p ! 关共兩 ᐉ 兩 + p兲 ! 兲兴其1/2.
ingly becoming obsolete: black-and-white film photography. One of the characteristics of the solution is that the ampli-
This method consists of taking a photograph of a computer tude has a purely radial dependence. When ᐉ ⫽ 0 and p = 0
printout of the pattern. The image on the film 共i.e., the nega- the beams have a characteristic single-ringed “doughnut”
tive兲 is the grating. If the pattern of the grating is composed shape, with the radius of the doughnut given by w冑ᐉ / 2. The
of two staggered gratings, then the first-order diffracted radius of curvature of the wave front and the Gouy phase are
beams will closely resemble the N = 1 Hermite-Gauss beam, the same as in the high-order rectangular solutions.
as shown in the bottom panel of Fig. 3共a兲. The most interesting term is the one containing the phase
By using the grating in one of the arms of the interferom- ᐉ␪. Ignoring for the moment the correction due to the radius
eter 共G in Fig. 1兲 one can interfere a beam in a high-order of curvature of the phase front and the Gouy phase 关or
mode with the original beam in the zero-order mode. The equivalently assuming R共z兲 → ⬁兴, the phase of the wave is
interference pattern consists of a staggered set of fringes, as given by
shown in Fig. 2共c兲. The experiment also demonstrates the
principle of holography: the interference pattern is a recon-
struction of the staggered grating 共i.e., the hologram兲. A low- ␺共␳, ␪,z兲 = kz + ᐉ ␪ . 共9兲
cost solution for making the grating is to reduce the com-
puter printout of Fig. 3共a兲 with a photocopy machine to about Points of constant phase 共i.e., the wave front兲 form a helix of
6 mm in size, and make the last photocopy onto transparency pitch ␭, as shown in Fig. 4 for the case ᐉ = 1. For ᐉ ⬎ 1 the
paper. wave front consists of ᐉ intertwined helices. In a given trans-
verse plane the phase depends on ␪. At ␳ = 0 the phase is
IV. HIGH-ORDER LAGUERRE-GAUSS SOLUTIONS undefined. At that point there is a phase singularity or
vortex,14 and the amplitude of the wave is zero.
The physics of high-order modes gets much more interest-
These beams are easily produced with a forked diffraction
ing when we consider the solutions of the paraxial equation
grating.15 Figure 3共b兲 shows a charge-1 forked diffraction
in cylindrical coordinates. This solution is not easily found in
grating. The pattern shown is a graph of the binary valued
the literature,13 but perhaps it is too mathematical to cover it
interference pattern of a Laguerre-Gauss beam with ᐉ = 1 and
in an undergraduate optics course. The solution in cylindrical
coordinates 共␳ = 冑x2 + y 2 , ␪ = tan−1共y / x兲 , z兲 is given by
a noncollinear plane wave. If the propagation vectors of the
wave of order ᐉ and a plane wave of order zero are 共0 , 0 , k兲
E p,ᐉ共␳, ␪,z兲 and 共k sin ␤ , 0 , k cos ␤兲, respectively, then the phase differ-
A 冑2␳
冉 冊 冉 冊 ᐉ ence between them at the z = 0 plane is
2␳2 −␳2/w共z兲2 ik␳2/关2R共z兲兴 iᐉ␪ i␾共z兲
= Lᐉp e e e e .
w共z兲 w共z兲 w共z兲2
␺共x,y,0兲 = ᐉ tan−1共y/x兲 − xk sin ␤ . 共10兲
共8兲
The solutions contain the associated Laguerre function Lᐉp, The binary forked grating of charge q = ᐉ is the filled contour
and therefore called Laguerre-Gauss modes. The subindices graph of the function

f共x,y兲 = 再 1,
0,
2m␲ + ⑀␲/2 艋 ␺共x,y,0兲 ⬍ 共2m + 1兲␲ − ⑀␲/2,
共2m + 1兲␲ − ⑀␲/2 艋 ␺共x,y,0兲 ⬍ 共2m + 2兲␲ + ⑀␲/2,
共11兲

358 Am. J. Phys., Vol. 74, No. 4, April 2006 Apparatus and Demonstration Notes 358

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
where m is an integer. One can set a graphing routine to fill reproduce because of the low efficiency in producing
with white where f共x , y兲 = 1, and black where f共x , y兲 = 0. The Laguerre-Gauss beams with the forked amplitude gratings.
ratio of the white to black thickness is given by 共1 − ⑀兲 / 共1 The recent report of a low-cost phase plate26 holds promise
+ ⑀兲. The pattern of Fig. 3共b兲 was made with ⑀ = 0.11. for implementing a feasible angular momentum transfer
One can use the interference setup of Fig. 1 to investigate demonstration in the undergraduate laboratory.
the peculiar phase of Laguerre-Gauss beams. Similar to the The orbital angular momentum carries over to the photon
discussion in the preceding section, we place a charge-1 level when the photons are propagating in that mode of the
forked diffraction grating in one of the arms of the interfer- light. This is an interesting but challenging concept to grasp.
ometer. In general, when a zero-order input beam is incident We must imagine that a single photon has a spatial extension
on a grating of charge q, the nth diffracted order is a in the transverse direction in order to carry the phase prop-
Laguerre-Gauss beam with ᐉ = qn 共see also Ref. 16兲. In the erties of a Laguerre-Gauss mode, and thus orbital angular
arm with the q = 1 forked grating we steer the first order 共i.e., momentum. The total angular momentum of a photon can be
with ᐉ = 1兲 to interfere with the zero order mode going expressed as18,22
through the other arm. When the reference beam is unex- L=␴ប + ᐉ ប, 共12兲
panded and noncollinear, the interference pattern is a recre-
ation of the forked hologram, as shown in Fig. 2共e兲. When where ␴ = ± 1 represents the spin angular momentum of the
the unexpanded reference beam is collinear with the one in photon.
the Laguerre-Gauss mode, the interference pattern consists of
an off-axis “blob” that gyrates about the central spot when VI. CONCLUSIONS
the path difference between the two interferometer arms is
The topic of Gaussian beams provides students with the
varied 共see, for example, Ref. 17兲. If the reference beam is
fundamentals for understanding the physics of laser beams
expanded, then the interference pattern has a unique spiral
and their propagation. Given the wide-spread use of lasers
shape, as shown in Fig. 2共f兲.
today, this material should be an essential part of a course on
There are other interesting aspects of high-order modes of
optics. The coverage of high-order Gaussian modes serves to
Gaussian beams that can be studied in upper-level under-
deepen the discussion of light waves, and to underscore the
graduate laboratory projects. One is the conversion between
key components of the wave function: amplitude and phase.
Hermite-Gauss and Laguerre-Gauss modes using astigmatic
The discussion of Laguerre-Gauss beams is necessary for
mode converters,10,11,18 which also constitutes an important
introducing students to the orbital angular momentum of the
application of the Gouy phase. Other interesting solutions to
light.
the wave equation are the “diffraction-free” Bessel beams19
We have successfully introduced this new material to un-
and the solution of the wave equation in elliptical-cylindrical
dergraduates in our optics course. As presented in this article,
coordinates, which is expressed in terms of Mathiew
we expanded the treatment of Gaussian beams to include
functions.20
beams in high-order modes, especially the Laguerre-Gauss
solutions of the wave equation. This discussion was pre-
V. ORBITAL ANGULAR MOMENTUM sented early in course, in a chapter on light waves. Orbital
angular momentum was introduced more formally later in
As mentioned before, Laguerre-Gauss beams carry orbital
the course, after a chapter on polarization. This way we
angular momentum. This is different from spin angular mo-
could cover under the unifying theme of momentum, the
mentum, which is associated with the circular polarization of
linear momentum of the light and both forms of angular
the light. Orbital angular momentum has been discussed ex-
momentum, spin and orbital. The phase of Gaussian beams
tensively in the literature.21,22 It can be understood intuitively
was revisited in the chapter on interference. The topic also
by analyzing the wave front of Fig. 4. In free space the local
provided interesting exercises, such as the calculation of in-
momentum of the beam is normal to the surface of the wave
terference patterns that could be compared to images of real
front. Since this momentum has an azimuthal component,
patterns, like those of Fig. 2. The experiments discussed in
then there is a finite longitudinal component of the angular
this article were used in the laboratory component of the
momentum. A rigorous derivation shows that the total angu-
course or in class demonstrations.
lar momentum of the beam along the propagation direction is
proportional to ᐉ.18,21,22
There have been numerous demonstrations of the transfer ACKNOWLEDGMENTS
of orbital angular momentum to matter. All of the demonstra- The authors thank P.R. Crawford, P.J. Haglin, V. Matos,
tions involve optical tweezers. Microscopic objects 共e.g., M.W. Pysher, H.I. Sztul, and R.E. Williams for their contri-
micron-sized latex spheres兲 can be trapped in the high- butions to our laboratory experience at Colgate University,
intensity region of the doughnut and carried along the ring and C.H. Holbrow and J. Noe for help and useful discus-
by the tilted wave front.23 In this case one can think that the sions.
wave front is doing the same job that a rotating drill bit does
to the filings after shaving them off the drilled material. It is APPENDIX: INTERFEROMETER ALIGNMENT
important to note that the light is not traveling in a helical
trajectory. The helical surface is the mathematical locus of Aligning the Mach Zehnder interferometer requires a sys-
points of equal phase. Absorptive particles have been shown tematic approach. Here we describe a method that produces
to rotate by the transfer of angular momentum,24 and larger an excellent alignment with not much frustration. This align-
irregular objects have been set to rotate via light-induced ment also gets the beam very close to the condition of zero
torques.25 Unfortunately these experiments are not easy to path-length difference. The basis of our alignment is simple:

359 Am. J. Phys., Vol. 74, No. 4, April 2006 Apparatus and Demonstration Notes 359

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
Fig. 5. Procedure to align a laser beam to the rows of holes on an optical breadboard 共see text兲.

3
to make all beams travel in directions parallel to the tapped M. V. Berry, “Much ado about nothing: Optical dislocation lines 共phase
holes of an optical breadboard. We instruct our students to singularities zeros, vortices. . .兲,” in Singular Optics, edited by M. S. Sos-
kin 关Proc. SPIE 3487, 1–5 共1998兲兴.
learn this procedure, outlined below, as a basic skill in optics 4
E. Hecht, Optics 共Addison Wesley, Reading, 2002兲.
laboratory procedures. 5
F. L. Pedrotti and L. S. Pedrotti, Introduction to Optics 共Prentice-Hall,
Suppose that we wish to align a beam after it is reflected Upper Saddle River, NJ, 1993兲.
off a mirror. The alignment goes as follows: 6
A. M. Almeida, E. Nogeira, and M. Belseley, “Paraxial imaging: Gauss-
ian beams versus paraxial-spherical waves,” Am. J. Phys. 67, 428–433
共1兲 The beam is first aimed by eye to be parallel to the holes 共1999兲.
on the breadboard. We put pairs of screws 共with socket 7
P. W. Milonni and J. H. Eberly, Lasers 共Wiley, New York, 1988兲.
or knurled heads兲 along the same row of holes in two 8
P. Nachman, “Mach-Zehnder interferometer as an instructional tool,”
positions: # 1 共near兲 and # 2 共far兲. This is shown in “step Am. J. Phys. 63, 39–43 共1995兲.
9
1” of Fig. 5. A. E. Siegman, Lasers 共University Science Books, Mill Valley, 1986兲.
10
共2兲 An iris is put in position # 1 with its mount touching the M. W. Beijersbergen, L. Allen, H. E. L. O. van der Veen, and J. P.
Woerdman, “Astigmatic laser mode converters and transfer of orbital
two screws in that position. The iris is then adjusted so
angular momentum.” Opt. Commun. 96, 123–132 共1993兲.
that the beam goes through the iris, as shown in “step 2” 11
M. Padgett, J. Arlt, and N. Simpson, “An experiment to observe the
of Fig. 5. intensity and phase structure of Laguerre-Gaussian laser modes,” Am. J.
共3兲 The iris is now put in position # 2, with its mount touch- Phys. 64, 77–82 共1996兲.
12
ing the screws, as in position # 1. If the beam goes J. Brandenberger, Experiments in laser physics and spectroscopy for un-
through the iris, then it is aligned because from positions 13
dergraduates, Lawrence University, 2005.
# 1 to # 2 the iris has been translated along a line parallel E. Zauderer, “Complex argument Hermite-Gaussian and Laguerre-
to the holes on the breadboard. If the beam does not go Gaussian beams,” J. Opt. Soc. Am. 3, 465–469 共1986兲.
14
M. Brambilla, F. Battipede, L. A. Lugiato, V. Penna, F. Prati, C. Tamm,
through the iris in position # 2 then the beam must be and C. O. Weiss, “Transverse laser patterns. I. Phase singularity crystals,”
steered by tilting the mirror until the beam goes through Phys. Rev. A 43, 5090–5113 共1991兲.
the iris, as shown in “step 3” of Fig. 5. 15
N. R. Heckenberg, R. McDuff, C. P. Smith, and A. G. White, “Generation
共4兲 Iterate the previous two steps until the alignment of optical phase singularities by computer-generated holograms,” Opt.
converges. Lett. 17, 221–223 共1992兲.
16
G. F. Brand, “Phase singularities in beams,” Am. J. Phys. 67, 55–60
共1999兲.
The interferometer is set up by putting the components 17
E. J. Galvez, P. R. Crawford, H. I. Sztul, M. J. Pysher, P. J. Haglin, and R.
one by one. We first put a beam splitter and align it so that E. Williams, “Geometric phase associated with mode transformation of
the reflected part of the beam is parallel to the rows of holes optical beams bearing orbital angular momentum,” Phys. Rev. Lett. 90,
on the breadboard, as described above. We follow putting the 203901-1–4 共2003兲.
18
two mirrors one by one, aligning them so that the steered L. Allen, M. W. Beijersbergen, R. J. C. Spreeuw, and J. P. Woerdman,
beams are also parallel to the holes of the breadboard. One of “Orbital angular momentum of light and the transformation of Laguerre-
the mirrors is mounted on a translation stage. By this point Gaussian laser modes,” Phys. Rev. A 45, 8185–8189 共1992兲.
19
we must have the two beams intersecting in air. The next step J. Durnin, J. J. Miceli, Jr., and J. H. Eberly, “Diffraction-free beams,”
Phys. Rev. Lett. 58, 1499–1501 共1987兲.
is to put the second beam splitter so that the two beams meet 20
J. C. Gutierrez-Vega, M. D. Iturbe-Castillo, and S. Chavez-Cerda, “Al-
at its beam-splitting surface. We tilt this beam splitter so that ternative formulation for invariant optical fields: Mathieu beams,” Opt.
the reflected beam is aligned with the holes of the bread- Lett. 25, 1493–1495 共2000兲.
21
board. Last, we adjust the translation stage until the two L. Allen, M. J. Padgett, and M. Babiker, “The orbital angular momentum
beams overlap. of light,” Progress in Optics XXXIX 共Elsevier Science B.V., New York,
1999兲.
22
Optical Angular Momentum, edited by L. Allen, S. M. Barnett, and M. J.
1
M. J. Padgett and L. Allen, “Light with a twist in its tail,” Contemp. Padgett 共Institute of Physics, Bristol, 2003兲.
23
Phys. 41, 275–285 共2000兲. J. E. Curtis and D. E. Grier, “Structure of optical vortices,” Phys. Rev.
2
J. F. Nye, Natural Focusing and Fine Structure of Light: Caustics and Lett. 90, 133901 共2003兲.
24
Wave Dislocations 共Institute of Physics, Bristol, 1999兲. H. He, M. E. J. Friese, N. R. Heckenberg, and H. Rubinsztein-Dunlop,

360 Am. J. Phys., Vol. 74, No. 4, April 2006 Apparatus and Demonstration Notes 360

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission
“Direct observation of transfer of angular momentum to absorptive par- equivalence of spin and orbital angular momentum of light: an optical
ticles from a laser beam with a phase singularity,” Phys. Rev. Lett. 75, spanner,” Opt. Lett. 22, 52–54 共1997兲.
826–829 共1995兲. 26
C. Rotschild, S. Zommer, S. Moed, O. Hershcovitz, and S. G. Lipson,
25
N. B. Simpson, K. Dholakia, L. Allen, and M. J. Padgett, “Mechanical “Adjustable spiral phase plate,” Appl. Opt. 43, 2397–2399 共2004兲.

361 Am. J. Phys., Vol. 74, No. 4, April 2006 Apparatus and Demonstration Notes 361

Downloaded 24 Dec 2012 to 138.26.31.3. Redistribution subject to AAPT license or copyright; see http://ajp.aapt.org/authors/copyright_permission

You might also like