This article was downloaded by: [FU Berlin]

On: 30 April 2015, At: 23:21

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Modern Optics

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tmop20

Quantum theory of light diffraction

a a a a a
Xiang-Yao Wu , Bai-Jun Zhang , Jing-Hai Yang , Li-Xin Chi , Xiao-Jing Liu , Yi-Heng Wu
a a a b c
, Qing-Cai Wang , Yan Wang , Jing-Wu Li & Yi-Qing Guo
a
Institute of Physics, Jilin Normal University , Siping 136000, China
b
Institute of Physics, Xuzhou Normal University , Xuzhou 221000, China
c
Institute of High Energy Physics , PO Box 918(3), Beijing 100049, China
Published online: 14 Oct 2010.

To cite this article: Xiang-Yao Wu , Bai-Jun Zhang , Jing-Hai Yang , Li-Xin Chi , Xiao-Jing Liu , Yi-Heng Wu , Qing-Cai Wang ,
Yan Wang , Jing-Wu Li & Yi-Qing Guo (2010) Quantum theory of light diffraction, Journal of Modern Optics, 57:20, 2082-2091,
DOI: 10.1080/09500340.2010.521593

To link to this article: http://dx.doi.org/10.1080/09500340.2010.521593

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Modern Optics
Vol. 57, No. 20, 20 November 2010, 2082–2091

Quantum theory of light diffraction

Xiang-Yao Wua*, Bai-Jun Zhanga, Jing-Hai Yanga, Li-Xin Chia, Xiao-Jing Liua, Yi-Heng Wua,
Qing-Cai Wanga, Yan Wanga, Jing-Wu Lib and Yi-Qing Guoc
a
Institute of Physics, Jilin Normal University, Siping 136000, China; bInstitute of Physics, Xuzhou Normal University,
Xuzhou 221000, China; cInstitute of High Energy Physics, PO Box 918(3), Beijing 100049, China
(Received 23 March 2010; final version received 31 August 2010)

At present, the theory of light diffraction only has the simple wave-optical approach. In this paper, we study light
diffraction with the relativistic quantum theory approach. We find that the slit length, slit width, slit thickness
and wavelength of light affect the diffraction intensity and form of diffraction pattern. However, the effect of slit
thickness on the diffraction pattern cannot be explained by wave-optical approach, but it can be explained in
quantum theory. We compare the theoretical results with single- and multiple-slits experimental data, and find
the theoretical results are in accordance with the experimental data. In addition, we give some theory predictions.
We think all new predictions will be tested by the light diffraction experiment.
Downloaded by [FU Berlin] at 23:21 30 April 2015

Keywords: quantum theory; light diffraction

1. Introduction description needs a quantum theory approach. In 1924,

It is known that the non-classical phenomena of two Epstein and Ehrenfest first studied light diffraction
photon interference [1] and two-photon ghost diffrac- with the old quantum theory, i.e. the quantum
tion and imaging [2,3] have classical counterparts. The mechanics of the correspondence principle, and
classical two-photon interference of light was first obtained a result identical to that obtained using
discovered in the pioneering experiments by Hanbury classical optics [11]. Here, in this work, we study the
Brown and Twiss [4] and since then it has been single-slit and multiple-slit diffraction of light from the
observed with various sources, including pseudo- relativistic quantum theory of photons approach. In
thermal ones [5], true thermal ones [6], and coherent view of quantum theory, the light has the nature of a
ones [7]. Somewhat later, ghost imaging with classical wave, and the wave is described by the wave function.
light was demonstrated, both in the near-field and far- As the wave equation we study has the character of a
field domains [8–10]. The present optical imaging vector, we choose wave function ~ð~r, tÞ to describe the
technologies, such as optical lithography, have reached wave. The wave function ~ð~r, tÞ can be calculated with
a spatial resolution in the sub-micrometer range, which a relativistic wave equation and it also has a statistical
comes up against the diffraction limit due to the meaning, i.e. j ~ð~r, tÞj2 can be explained as the photon’s
wavelength of light. However, the guiding principle of probability density at the definite position. In light
such technology is still based on the classical diffrac- diffraction, because the diffraction intensity I is
tion theory established by Fresnel, Kirchhoff and directly proportional to j ~ð~r, tÞj2 , we can obtain the
others more than a hundred years ago. Recently, the diffraction intensity by calculating the light wave
use of quantum-correlated photon pairs (biphotons) to function ~ð~r, tÞ distributed on the display screen, and
overcome the classical diffraction limit was proposed the light wave functions can be divided into three
and attracted much attention. Obviously, quantum areas. The first area is the incident area, where the
theory approaches are necessary to explain the photon wave function is a plane wave. The second area
diffraction-interference of the quantum-correlated is the slit area, where the light wave function can be
multi-photon state. As is well known, Huygens’ and calculated by the quantum wave equation of light. The
Kirchhoff’s theory, has been successfully applied to third area is the diffraction area, where the light wave
classical optics, and has yielded good agreement with function can be calculated by Kirchhoff’s law. For
many experiments. However, light interference and multiple-slit diffraction, we can obtain the total
diffraction are quantum phenomena, and their full diffraction wave function by the superposition of the

*Corresponding author. Email: [email protected]

ISSN 0950–0340 print/ISSN 1362–3044 online

ß 2010 Taylor & Francis
DOI: 10.1080/09500340.2010.521593
http://www.informaworld.com
 Journal of Modern Optics 2083

where c0 is the slit thickness. We can get the

time-dependent relativistic wave equation in the slit
(Vðx, y, zÞ ¼ 0), it is
@ ~
ih ð~r, tÞ ¼ chr  ~ð~r, tÞ, ð4Þ
@t
by derivation of Equation (4) about time t and
multiplying both sides by i", we have
Figure 1. The single-slit geometry, a is the width and b is the
@2 ~ @
length of the slit. ðihÞ2 2
ð~r, tÞ ¼ chr  ih ~ð~r, tÞ: ð5Þ
@t @t
Substituting Equation (4) into Equation (5),
we have
diffraction wave function of every slit. In the following,
we will calculate these wave functions. @2 ~ h i
r
ð~, tÞ ¼ c 2
rðr  ~ð~r, tÞÞ  r2 ~ð~r, tÞ , ð6Þ
The paper is organized as follows. In Section 2 we @t2
calculate the light wave function in the single-slit with
the quantum theory approach. In Section 3 we where the formula r  r  B~ ¼ rðr  BÞ ~  r2 B. ~
p"ffiffiffi0 From
~
[11], the photon wave function is ð~r, tÞ ¼ 2 ðEð~ ~ r, tÞþ
calculate the light wave function in the diffraction ~
icBð~r, tÞÞ, we have
Downloaded by [FU Berlin] at 23:21 30 April 2015

area with Kirchhoff’s law. Section 4 is the multiple-slit

diffraction wave function. Section 5 is the numerical r  ~ð~r, tÞ ¼ 0, ð7Þ
result. Section 6 is a summary of results and the
conclusions. from Equations (6) and (7), we have
 2 
@ 2 2 ~
c r ð~r, tÞ ¼ 0: ð8Þ
@t2
2. Quantum approach of light single-slit diffraction
Equation (8) is the same as the classical wave
In an infinite plane, we consider a single-slit, its width a equation of light. Here, it is a quantum wave equation
and length b are shown in Figure 1. The x-axis is along of light, since it is obtained from the relativistic wave
the slit length and the y-axis is along the slit width. Equation (2), and it satisfied the new quantum
In the following, we calculate the light wave function in boundary condition: when ~ð~r, tÞ ! 0, Vð~r Þ ! 1.
the single-slit with a relativistic wave equation. At time It is different from the classical boundary condition.
t, we suppose that the incident plane wave travels along When the photon wave function ~ð~r, tÞ changes
the z axis. It is with determinate frequency !, the wave function of
photon can be written as
~ hi ð pzEtÞ
~0 ðz, tÞ ¼ Ae
X i ~ð~r, tÞ ¼ ~ð~r Þei!t : ð9Þ
¼ Aj  ehð pzEtÞ e~j
j
X i
Substituting Equation (9) into Equation (8),
¼ 0j  ehEt e~j , ð1Þ we can get
j
i @2 ~ð~r Þ @2 ~ð~r Þ @2 ~ð~r Þ 42 ~
where 0j ¼ Aj  ehpz , j ¼ x, y, z and A~ is a constant þ þ þ 2 ð~r Þ ¼ 0, ð10Þ
vector. The time-dependent relativistic wave equation @x2 @y2 @z2 
of light is [12] and the wave function satisfies boundary conditions
@ ~ ~ð0, y, zÞ ¼ ~ðb, y, zÞ ¼ 0, ð11Þ
i
h hr  ~ð~r, tÞ þ V ~ð~r, tÞ,
ð~r, tÞ ¼ c ð2Þ
@t
where c is light velocity. From Equation (2), we can ~ðx, 0, zÞ ¼ ~ðx, a, zÞ ¼ 0: ð12Þ
find the light wave function ~ð~r, tÞ ! 0 when
Vð~r Þ ! 1. The potential energy of light in the The photon wave function ~ð~r Þ can be written
single-slit is
~ð~r Þ ¼ r Þ~ex þ y ð~r Þ~
x ð~ ey þ r Þ~
z ð~ ez
Vðx, y, zÞ ¼ 0 0  x  b, 0  y  a, 0  z  c 0 X
¼ j ð~ ej ,
r Þ~ ð13Þ
¼1 otherwise, ð3Þ j¼x, y,z
 2084 X.-Y. Wu et al.

where j is x, y or z. Substituting Equation (13) into in the slit

Equations (10), (11) and (12), we have the component X
equation ~ðx, y, z; tÞ ¼ j ðx, y, z, tÞ~
ej
j¼x, y,z

@2 j ð~r Þ @2 j ð~r Þ @2 j ð~r Þ 42 X X

1
16Aj
þ þ þ 2 rÞ
j ð~ ¼ 0: ð14Þ ¼
@x2 @y2 @z2  j¼x, y,z m,n¼0
ð2m þ 1Þð2n þ 1Þ2
ð2n þ 1Þx ð2m þ 1Þy
 sin sin
b a
j ð0, y, zÞ ¼ j ðb, y, zÞ ¼ 0, ð15Þ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
42 ð2nþ1Þ2 2 ð2mþ1Þ2 2
i   z
e 2 b2 a2
ei!t e~j : ð23Þ
j ðx, 0, zÞ ¼ j ðx, a, zÞ ¼ 0: ð16Þ We can consider the case of a limit, i.e. the slit
length b is infinity, and the Equations (8) and (10)
The partial differential Equation (14) can be solved become
by the method of separation of variables.  2 
@2 ~ 2 @ @2 ~
j ðx, y, zÞ ¼ Xj ðxÞYj ð yÞZj ðzÞ: ð17Þ ð y, z, tÞ  c þ ð y, z, tÞ ¼ 0, ð24Þ
@t2 @y2 @z2
From Equations (14), (15), (16) and (17), we can get
the general solution of Equation (14) @2 ~ð y, zÞ @2 ~ð y, zÞ 42 ~
Downloaded by [FU Berlin] at 23:21 30 April 2015

þ þ 2 ð y, zÞ ¼ 0, ð25Þ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi @y2 @z2 
X nx my i 422 n2b2 2 ma222 z
j ðx, y, zÞ ¼ Dmnj sin sin e , we can easily obtain the light wave function in the
mn
b a single-slit when b ! 1
ð18Þ X
~ð y, z; tÞ ¼ j ðx, y, z, tÞ~
ej
since the wave functions are continuous at z ¼ 0, j¼y,z
we have X X
1
4Aj ð2m þ 1Þy
¼ sin
~0 ðx, y, z; tÞ jz¼0 ¼ ~ðx, y, z; tÞ jz¼0 ; ð19Þ j¼x, y,z m¼0
ð2m þ 1Þ a
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
42 ð2mþ1Þ2 2
or, equivalently, i  z
e 2 a2
ei!t e~j : ð26Þ
0j ðx, y, zÞ jz¼0 ¼ j ðx, y, zÞ jz¼0  ð j ¼ x, y, zÞ: ð20Þ

From Equations (1), (18) and (20), we obtain the

3. The wave function of light diffraction
coefficient Dmnj by Fourier transform
In Section 2, we calculated the photon wave function in
ðaðb the slit. In the following, we will calculate the diffrac-
4 n m
Dmnj ¼ Aj sin sin dd tion wave function. We can calculate the wave function
ab 0 0 b a
in the diffraction area. From the slit wave function
16Aj
¼ m, n, odd component j ð~r, tÞ, we can calculate its diffraction
mn2
wave function component j ð~r, tÞ by Kirchhoff’s law.
¼ 0 otherwise, ð j ¼ x, y, zÞ ð21Þ It can be calculated by the formula [13]
ð ikr    
substituting Equation (21) into Equation (18), we have 1 e ! 1 r~
j ð~r, tÞ ¼  n  50 j þ ik  j ds:
X
1 4 s0 r r r
16Aj ð2n þ 1Þx
j ðx, y, zÞ ¼ 2
sin ð27Þ
m,n¼0
ð2m þ 1Þð2n þ 1Þ b
ð2m þ 1Þy The total diffraction wave function is
 sin X
a ~ r, tÞ ¼
ð~ j ð~r, tÞ~
ej ,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð28Þ
42 ð2nþ1Þ2 2 ð2mþ1Þ2 2 j¼x, y,z
i   z
2 b2 a2
e , ð j ¼ x, y, zÞ
the diffraction area is shown in Figure 2, where k ¼ 2 
ð22Þ
is the wave vector, s0 is the area of the single-slit, r~0 is
and substituting Equation (22) into Equations (9) the position of a point on the surface ðz ¼ c0 Þ, P is an
and (13), we can obtain the photon wave function arbitrary point in the diffraction area, and the ! n is a
 Journal of Modern Optics 2085
!
where is the angle between k2 and the z axis.
Substituting Equations (32) and (33) into Equation
(31) gives
eikR i!t X 1 X 1
16Aj
j ðx,y,z; tÞ ¼  e
4R m¼0 n¼0
ð2m þ 1Þð2n þ 1Þ2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
42 ð2nþ1Þ 2 ð2mþ1Þ 2 0
 ei 2 ð b Þ ð a Þ c
2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   
4 42 ð2n þ 1Þ 2 ð2m þ 1Þ 2
 i  
2 b a
Figure 2. The diffraction area of single-slit.
3
 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
þ ik  cos2   sin2 5
unit vector, which is normal to the surface of the R
single-slit. From Figure 2, we have ðb ð
ik sin x0 ð2n þ 1Þ 0 0 a ik sin y0
!  e sin x dx e
R 0 b 0
r ¼ R   r~0
R ð2m þ 1Þ 0 0
 sin y dy : ð34Þ
r~ 0 a
 R   r~
r
Downloaded by [FU Berlin] at 23:21 30 April 2015

Substituting Equation (34) into Equation (28),

!
k2 one can get
¼ R   r~0 , ð29Þ
k
~ eikR i!t X X 1 X 1
16Aj
ðx,y,z; tÞ ¼  e
then, 4R ð2m þ 1Þð2n þ 1Þ2
j¼s, y,z m¼0 n¼0
r~ 0 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
eikr eikðRr~r Þ 42 ð2nþ1Þ 2 ð2mþ1Þ 2 0
¼  ei 2 ð b Þ ð a Þ c
r R  rr~  r~0 2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   
! 0 42 ð2n þ 1Þ 2 ð2m þ 1Þ 2
eikR ei k2 ~r  4i  
 ðj~r 0 j  RÞ, ð30Þ 2 b a
R 3
 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
where k~2 ¼ k rr~. Substituting Equations (22), (29) and 1
(30) into Equation (27), one can obtain þ ik  cos2   sin2 5
R
ð ! 0
eikR i!t ðb ð
j ð~r, tÞ ¼  e ei k2 ~r 0 ð2n þ 1Þ 0 0 a ik sin y0
4R s0  eik sinx sin x dx e
0 b 0
X1 X 1
16Aj
 ð2m þ 1Þ 0 0
ð2m þ 1Þð2n þ 1Þ2  sin y dy e~j : ð35Þ
m¼0 n¼0 a
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
42 ð2nþ1Þ 2 ð2mþ1Þ 2 0 ð2n þ 1Þ Equation (35) is the total diffraction wave function
 ei 2 ð b Þ ð a Þ c sin
b in the diffraction area. From the wave function, we can
0 ð2m þ 1Þ 0 obtain the diffraction intensity I on the display screen,
 x sin y
a
2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi we have  2
    ~ 
42 ð2n þ 1Þ 2 ð2m þ 1Þ 2 I / ðx, y, z; tÞ : ð36Þ
 4i  
2 b a
3
! 4. Multiple-slit diffraction wave function of light
! ! n R5 0 0
þi!n  k2  dx dy : ð31Þ From Equation (23), in the first slit, the photon wave
R2
function ~1 ðx, y, z; tÞ is
!
Assume that the angle between k2 and the x axis X X
1
  ~1 ðx, y, z; tÞ ¼ 16Aj
(y axis) is 2   ( 2  ), and ðÞ is the angle between
! j¼x, y,z m,n¼0
ð2m þ 1Þð2n þ 1Þ2
k2 and the surface of yz (xz), then we have
ð2n þ 1Þx ð2m þ 1Þy
k2x ¼ k sin , k2y ¼ k sin , ð32Þ  sin sin
b a
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
42 ð2nþ1Þ2 2 ð2mþ1Þ2 2
! ! i   z
n  k2 ¼ k cos , ð33Þ e 2 b2 a2
ei!t e~j : ð37Þ
 2086 X.-Y. Wu et al.
ikR X X 1 X 1
16Aj
~ 2 ðx, y, z;tÞ ¼  e ei!t

4R j¼x, y,z m¼0 n¼0
ð2m þ 1Þð2n þ 1Þ2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
42 ð2nþ1Þ 2 ð2mþ1Þ 2 0
 ei 2 ð b Þ ð a Þ c
2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   
42 ð2n þ 1Þ 2 ð2m þ 1Þ 2
 4i  
2 b a
3
 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Figure 3. Multiple-slit geometry with a the single slit width, 1
b the slit length and d the distance between the two slits. þ ik  cos2   sin2 5
R
ðb
0 ð2n þ 1Þ 0 0
From Figure 3, in the second slit, when we make  eiksin x sin x dx
the coordinate translations: 0 b
ð 2aþd
0 ð2m þ 1Þ 0 0
x0 ¼ x  eik sin y sin y dy e~j : ð42Þ
aþd a
y0 ¼ y  ða þ d Þ
 
z0 ¼ z, ð38Þ ikR X X 1 X 1
~ N ðx,y, z; tÞ ¼  e ei!t 16Aj
we can obtain the second slit photon wave function by 
4R j¼x,y,z m¼0 n¼0
ð2m þ 1Þð2n þ 1Þ2
the first slit photon wave function. It is pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
42 ð2nþ1Þ 2 ð2mþ1Þ 2 0
 ei 2 ð b Þ ð a Þ c
Downloaded by [FU Berlin] at 23:21 30 April 2015

X X
1
16Aj 2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
~2 ðx,y,z; tÞ ¼    
ð2m þ 1Þð2n þ 1Þ2 42 ð2n þ 1Þ 2 ð2m þ 1Þ 2
j¼x,y,z m,n¼0  4i  
2 b a
ð2n þ 1Þx ð2m þ 1Þ½y  ða þ d Þ 3
 sin sin  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
b a 1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi þ ik  cos2   sin2 5
42 ð2nþ1Þ2 2 ð2mþ1Þ2 2 R
i   z
e 2 b2 a2
ei!t e~j : ð39Þ ðb
0 ð2n þ 1Þ 0 0
Similarly, we can also obtain the Nth photon wave  eiksin x sin x dx
0 b
function. It is ð ðN1Þðaþd Þþa
0 ð2m þ 1Þ 0 0
X X 1
16Aj ð2n þ 1Þx  eiksin y sin y dy e~j :
~N ðx, y, z;tÞ ¼ sin ðN1Þðaþd Þ a
ð2m þ 1Þð2n þ 1Þ 2 b
j¼x, y,z m,n¼0
ð43Þ
ð2m þ 1Þ½ y  ðN  1Þða þ dÞ The total diffraction wave function for the
 sin
a Nth-slit is
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
42 ð2nþ1Þ2 2 ð2mþ1Þ2 2 z ~ y, z; tÞ ¼ 
~ 1 ðx, y, z; tÞ þ 
~ 2 ðx, y, z; tÞ þ      
i ðx,
e 2 b2 a2
ei!t e~j : ð40Þ
~ N ðx, y, z; tÞ:
þ ð44Þ
With Kirchhoff’s law, similar to Equation (35), we
From Equation (44), we can obtain the diffraction
can get the light diffraction wave function in every slit,
intensity I on the display screen for the Nth-slit,
these are
so we have ~ y, z; tÞj2 : ð45Þ
ikR X X 1 X 1
16Aj I / jðx,
~ 1 ðx, y, z;tÞ ¼  e ei!t
4R j¼x,y,z m¼0 n¼0
ð2m þ 1Þð2n þ 1Þ2 When b ! 1, we can get the single-slit light
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi diffraction wave function
42 ð2nþ1Þ 2 ð2mþ1Þ 2 0
 ei 2 ð b Þ ð a Þ c ~ b!1 ð y, z; tÞ

2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
    eikR i!t X X 1
4 42 ð2n þ 1Þ 2 ð2m þ 1Þ 2 ¼ e
4Aj 42 ð2mþ1Þ 2 0
ei 2 ð a Þ c
 i   4R ð2m þ 1Þ
2 b a j¼x, y,z m¼0
2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
3    
 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 42 ð2m þ 1Þ 2 1
1  4i  þ ik 
þ ik  cos2   sin 5 2
2 a R
R
3
ðb ð qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0 ð2n þ 1Þ 0 0 a iksin y0  cos2   sin2 5
 eik sinx sin x dx e
0 b 0
ð2m þ 1Þ 0 0 ða
 sin y dy e~j : 0 ð2m þ 1Þ 0 0
a
ð41Þ  eik sin y sin y dy e~j : ð46Þ
0 a
 Journal of Modern Optics 2087

From Equation (46), we can obtain the diffraction where

intensity Ib!1 on the display screen for single-slit
when b ! 1. It is a sin ða þ d Þ sin
¼ , ¼ , ð49Þ
 
~ b!1 ð y, z, tÞj2 :
Ib!1 / j ð47Þ is the diffraction angle, a is the width of slit, d is the
distance from the first slit to the second slit, and n is the
In [11], the authors had first studied light diffrac-
number of slits.
tion with the old quantum theory, i.e. the quantum
Figure 4(a)–( f ) shows the diffraction patterns from
mechanics of the correspondence principle. They had
two, three, four, five, six and seven slits with
considered a quantum of light entering a three-
 ¼ 6:328  10 7 m, the slit width a ¼ 0:88  10 4 m,
dimensional crystal lattice, and calculated the light
a þ d ¼ 3:52  10 4 m (d is the distance between two
momentum loss after collision, with the quantization
slits), and the distance between slit and display screen
condition and the light quantum momentum formula
R ¼ 4.572 m. In Figure 4(a)–( f ), the solid curve is
p ¼ h. They obtained the relation between the deflect-
our theoretical result, and the dotted curve is the
ing angle of the light quantum collision with the lattice
result of Equation (48), i.e. the diffraction data.
and the lattice period. By analyzing this, the authors
In Figure 4(a)–( f ), we use the same experimental
given the expression of electronic density distributed
parameters in our calculation, and the theoretical input
on the grating
¼ Am sin 2mx
a . The coefficient Am could
parameters are: slit length b ¼ 3:52  10 4 m, and slit
be obtained by Fourier analysis. The diffraction
thickness c0 ¼ 0:88  10 4 m. From Figure 4(a), we
Downloaded by [FU Berlin] at 23:21 30 April 2015

spectrum intensity of the mth order is proportional

to A2m . Finally, they obtained the intensity formula of can find the calculation result is in accordance with
the diffraction spectrum, which was in complete experimental data. Since the ratio aþd a ¼ 4, we find the
agreement with the classical diffraction. The formula orders 4, 8, 12 . . . are missing. The conclusion is
gives a simple relation between diffraction intensity, accordance with classical optics.
slit width, wavelength of incident light and diffraction Figure 4(b)–( f ) are multiple slits diffraction pat-
angle. In this paper, we present the quantum theory of terns corresponding to slit number N¼3, 4, 5, 6, 7.
light diffraction using the framework of a relativistic We can find the calculation results are in accordance
quantum theory, and obtain the relation between the with experiment data, and there are N  2 secondary
diffraction intensity, slit length, slit width, slit thick- maxima and N  1 minima between the two principle
ness, light wavelength and diffraction angle. Through maxima. The conclusions are accordance with classical
calculation, we find the theoretical results are in optics.
accordance with the experimental data, and give Figure 5 shows the diffraction patterns for the
some new theory predictions. single slit. In the experiment, the light wavelength
 ¼ 6:328  10 7 m, the slit width a ¼ 1:76  10 4 m,
and the distance between the slit and display screen
R ¼ 4.572 m. In our calculation, we take the same
5. Numerical results
experimental parameters above, and the theoretical
The light diffraction experiment of single and multiple input parameters are: slit length b ¼ 4:0  10 4 m, the
slits was reported by H.F. Neiners in 1970 [14]. In the slit thickness c0 ¼ 1:1  10 6 m, and the diffraction
experiment [14], the optical system consisted of two angle ¼0.001 rad. In Figure 5, the solid curve is our
convex lenses, the focal length f, a diffraction screen of theoretical result and the dotted curve is the result of
slit length b and slit width a (b a) and a display Equation (48), i.e. the diffraction data. From Figure 5,
screen. The laser light source of wavelength  is placed we find the calculation result shows good agreement
on the focal plane of the first convex lens, the first lens with experimental data. We have compared the theo-
makes the light beams parallel and incident on the retical results with the experiment data above. In the
diffraction screen, the second lens is next to the following, we give some theoretical predictions.
diffraction screen, and the display screen is placed In the calculations, we take the light wavelength to
on the focal plane of second convex lens. In the be ¼6.328107 m, the distance R ¼ 4.572 m and
experiment, the diffraction patterns were shown by the diffraction angle ¼0.001 rad.
photographs, but experimental data were not given. Figure 6 is obtained by taking the single slit
The author found his results could be explained width as 5a, 10a and 20a (a ¼ 1.76  104 m) and
excellently by the classical theoretical formula. It is b ¼ 4.0  104 m, c0 ¼ 1:1  10 6 m. From Figure 6, we
find that when the slit width increases, the diffraction
sin2  sin2 N patterns become narrower, and the diffraction intensity
I ¼ I0 , ð48Þ
2 sin2  increases.
 2088 X.-Y. Wu et al.

(a) (b)

(c) (d )
Downloaded by [FU Berlin] at 23:21 30 April 2015

(e) (f)

Figure 4. The diffraction patterns from two, three, four, five, six and seven slits with a ¼ 0:88  10 4 m, a þ d ¼ 3:52  10 4 m,
b ¼ 4  0:88  10 4 m, c0 ¼ 0:88  10 4 m. The solid curve is our theoretical calculations and the dot curve is the result of
Equation (48).

In Figure 7, the slit width and slit length are equal i.e. a very small hole cannot produce diffraction
(a ¼ b), and this is obtained by taking the slit length phenomenon. This is because the incident light scatters
and slit width as , 3 and 5, and c0 ¼ 1:1  10 6 m. back completely when the size of slit is very small.
From Figure 7, we can find that when they increase, Figure 9 is obtained by taking the single slit length
the diffraction patterns become narrower, and the as 50b, 70b and infinity (b ¼ 4:0  10 4 m) and
diffraction intensity increases. a ¼ 1:76  10 4 m, c0 ¼ 1:1  10 6 m. From Figure 9,
In Figure 8, we obtain an important result: when we can obtain the following conclusions. (1) When the
a ¼ b  0:1, the total diffraction intensity is zero, slit length increases, the diffraction intensity increases.
 Journal of Modern Optics 2089

Figure 5. The diffraction patterns from single slit Figure 7. The diffraction patterns from single slit with
a ¼ 1:76  10 4 m, b ¼ 4:0  10 4 m and c0 ¼ 1:1 10 6 m. c0 ¼ 1:1  10 6 m. The solid, dash and dot curves correspond
The solid curve is our theoretical results and the dot curve is to a ¼ b ¼ , a ¼ b ¼ 3 and a ¼ b ¼ 5.
the result of Equation (48).
Downloaded by [FU Berlin] at 23:21 30 April 2015

Figure 6. The diffraction patterns from single slit with

b ¼ 4:0  10 4 m and c0 ¼ 1:1  10 6 m. The dash, solid and Figure 8. The diffraction patterns from single slit with
dot curves correspond to slit width 5a, 10a and c0 ¼ 1:1  10 6 m and a ¼ b  0:1. Their real intensity
20a(a ¼ 1:76  10 4 m), respectively. I should be multiplied by 102 and 103.

(2) When the slit length changes, the width of the and a ¼ 1:76  10 4 m, b ¼ 4:0  10 4 m, c0 ¼ 1:1
diffraction patterns does not change. 10 6 m. From Figure 11, we can obtain the follow-
Figure 10 is obtained by taking the single slit ing conclusions: (1) when the wavelength decreases, the
thickness as 100c0 , 1000c0 , 2000c0 and 3000c0 (c0 ¼ 1:1 total diffraction intensity increases and the diffraction
10 6 m) and a ¼ 1:76  10 4 m, b ¼ 4:0  10 4 m. patterns become narrow; (2) when the wavelength
From Figure 10, we can obtain the following conclu- decreases, the number of diffraction patterns increases.
sions: (1) when the slit thickness increases, the total Figure 12 is obtained by taking the double slit
diffraction intensity decreases; (2) when the slit thick- thickness as c0 , 10c0 and 50c0 (c0 ¼ 0:88  10 4 m) and
ness increases, the diffraction patterns are spread over a ¼ 0:88  10 4 m, b ¼ 4  0:88  10 4 m, a þ d ¼
a wide area. 3:52  10 4 m. From Figure 12, we can obtain
Figure 11 is obtained by taking the single wave- the following conclusions: (1) when the slit thickness
length as 10, 20 and 50 ( ¼ 6:328  10 7 m) increases, the diffraction intensity decreases;
 2090 X.-Y. Wu et al.

Figure 9. The diffraction patterns from single slit with Figure 11. The diffraction patterns from single slit with
a ¼ 1:76  10 4 m and c0 ¼ 1:1  10 6 m. The solid, dash a ¼ 1:76  10 4 m, b ¼ 4:0  10 4 m and c0 ¼ 1:1  10 6 m.
and dot curves curves correspond to slit length 50b0 , 70b0 and The solid, dash and dash-dot curves correspond to wave-
infinity (b0 ¼ 4:0  10 4 m), respectively. lengths 10, 20 and 50.
Downloaded by [FU Berlin] at 23:21 30 April 2015

Figure 10. The diffraction patterns from single slit with Figure 12. The diffraction patterns from double slit with
a ¼ 1:76  10 4 m and b ¼ 4:0  10 4 m. The dot, solid, a ¼ 0:88  10 4 m, b ¼ 4  0:88  10 4 m and a þ d ¼
dash-dot and dash curves correspond to slit thickness 100c0 , 3:52  10 4 m. The dot, solid and dash curves correspond
1000c0 , 2000c0 and 3000c0 (c0 ¼ 1:1  10 6 m), respectively. to slit thickness c0 , 10c0 and 50c0 .

(2) in classical optics, we know when the ratio aþd a ¼n 6. Conclusion

(n ¼ 1, 2, 3   ), the orders n, 2n, 3n, . . . are missing in In conclusion, we have studied the single-slit, double-
double slit diffraction. We find that when the slit slit and multiple-slit diffraction of light with the
thickness takes c0 , and the ratio aþd
a ¼ 4, the orders 4, 8, relativistic quantum mechanical approach. We give
12, . . . are missing. (3) When the slit thickness
the relation between diffraction intensity, slit length,
increases, such as 10c0 and 50c0 , we find the missing-
slit width, slit thickness, wavelength of light and
order phenomenon disappears.
From Figures 4 and 5, we can find our calculation diffraction angle. Our calculation results are in accor-
results are in accordance with the experimental data, dance with the experiment data of the single-slit,
the classical optics results and the old quantum theory double-slit and multiple-slit diffraction. In addition, we
results [13]. From Figures 6 to 12, we give some new study the effect of slit length, slit width, slit thickness,
predictions, which can be tested by light diffraction the wavelength of light on the diffraction intensity and
experiments. form of diffraction pattern. The effect of slit length and
 Journal of Modern Optics 2091

slit thickness on the diffraction pattern cannot be [5] Haner, A.B.; Isenor, N.R. Am. J. Phys. 1970, 38,
obtained in classical optics. In double slit diffraction, 748–751.
we find when the ratio aþd [6] Zhai, Y.-H.; Chen, X.-H.; Zhang, D.; Wu, L.-A. Phys.
a ¼ n (n¼1, 2, 3 . . .), the orders
Rev. A 2005, 72, 043805.
n, 2n, 3n, . . . are missing. When the slit thickness
[7] Pfleegor, R.I.; Mandel, L Phys. Rev. 1967, 159,
increases, the missing order phenomenon disappears.
1084–1088; Ou, Z.Y.; Gage, E.C.; Magill, B.E.;
In multiple slit diffraction, we find there are N  2 Mandel, L. Opt. Commun. 1988, 69, 1–5.
secondary maxima and N  1 minima between the two [8] Gatti, A.; Brambilla, E.; Bache, M.; Lugiato, L.A. Phys.
principle maxima, we think that all the new predictions Rev. Lett. 2004, 93, 093620; Ferri, F.; Magatti, D.;
in our work can be tested by light diffraction Gatti, A.; Bache, M.; Brambilla, E.; Lugiato, L.A. Phys.
experiments. Rev. Lett. 2005, 94, 183602.
[9] Bennink, R.S.; Bentley, S.J.; Boyd, R.W. Phys. Rev.
Lett. 2002, 89, 113601.
References [10] Scarcelli, G.; Valencia, A.; Shih, Y. Phys. Rev. A 2004,
70, 051802(R); DAngelo, M.; Valencia, A.; Rubin,
[1] Ghosh, R.; Mandel, L. Phys. Rev. Lett. 1987, 59, M.H.; Shih, Y. Phys. Rev. A 2005, 72, 013810.
1903–1905. [11] Epstein, P.S.; Ehrenfest, P. Proc. Nat. Acad. Sci. 1924,
[2] Strekalov, D.V.; Sergienko, A.V.; Klyshko, D.N.; Shih, 10, 133–139.
Y.H. Phys. Rev. Lett. 1995, 74, 3600–3603. [12] Smith, B.J.; Raymer, M.G. New J. Phys. 2007, 9, 414.
[3] Pittman, T.B.; Shih, Y.H.; Strekalov, D.V.; Sergienko, [13] Schwartz, M. Principles of Electrodynamics; Oxford
A.V. Phys. Rev. A 1995, 52, R3429–R3432. University Press: Oxford, 1972.
Downloaded by [FU Berlin] at 23:21 30 April 2015

[4] Hanbury Brown, R.; Twiss, R.Q. Nature 1956, 178, [14] Meiners, H.F. Physics Demonstration Experiments;
1046–1048. The Ronald Press: New York, 1970; Vol. II.