Results in Optics 12 (2023) 100439

Contents lists available at ScienceDirect

Results in Optics
journal homepage: www.sciencedirect.com/journal/results-in-optics

A miniature design of binary subtractor using 2-D photonic crystal structure

for high-speed applications
Sandip Swarnakar a, *, Katta Bhavan Kumar a, Kuruva Chavadi Yashwanth a, Yerravalli Saikiran a,
Naddi Venkata Rakesh a, Arjuna Muduli b, Santosh Kumar c
a
Photonics Lab, Department of Electronics and Communication Engineering, G. Pullaiah College of Engineering and Technology, Kurnool, Andhra Pradesh 518002, India
b
Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur 522302, Andhra Pradesh, India
c
Shandong Key Laboratory of Optical Communication Science and Technology, School of Physics Science and Information Technology, Liaocheng University, Liaocheng
252059, Shandong, China

A R T I C L E I N F O A B S T R A C T

Keywords: The expansion of the digital market is accelerating, and for faster communication, optical technologies have been
Photonic crystals waveguides developed. The binary subtractor is a combinational circuit that can compute the subtraction operations in
Interference principle arithmetic logic unit (ALU). The desired design is depicted as a 2-D structure with two T-shaped waveguides, and
FDTD
it is operates at a wavelength of 1550 nm. The design is verified and analyzed using finite-difference-time-
Photonic band gap
Contrast ratio
domain (FDTD) method. It is also verified using MATLAB software. The identical design has been computed
Insertion loss and simulated numerous times in this work, but with different lattice constants, refractive indices, and silicon rod
radii. In this study, photonic crystals are used to construct a binary subtractor that is both compact and capable of
producing a high contrast ratio (5.185 dB for difference and 17.78 dB for borrow). Moreover, the transmission
efficiency is achieved 89.3%, insertion loss is 0.49 dB, and the response time and bit rate are achieved by
0.0204 Ps and 49.01 Tbps respectively.

1. Introduction the speed of light has a velocity of 3 × 108 m/s; consequently, optical
devices have quick response and are also compact (Andalib and Gran­
In the beginning, the systems were constructed using vacuum diodes payeh, 2009; Calhoun et al., 2008). The devices are becoming quicker
as the main employed component (Abdullin and Morozo, 2013). Later and more complex as time goes on, but they do not meet the expecta­
PN junction diodes (Schneider and Strutt, 1959) were invented as a tions of people like fast switching devices (Jin and Wada, 2014); less
response to the enormous size of vacuum diodes as well as the significant delay time, compact size, etc., thus optical signaling is employed to
delay in transmission speed. These diodes are faster than vacuum diodes make the devices faster and smaller in size. One way to think of an
in transmitting data, and similar devices are being developed to fulfill optical fiber (Zhang et al., 2022) is that it is a type of dielectric wave­
the requirements of modern-day technology. Utilizing nanotechnology, guide (Li and Ho, 2003) that is designed to transmit signals at optical
all of the diodes, transistors, and digital circuitry (Zhang et al., 2007) are frequencies. In general, the waveguide refers to a device or a tube which
combined onto a single circuit called integrated circuit (IC). At this point in order to radiate energy, must either be bent or terminated. These
of time, these devices are utilized in various kinds of systems such as in devices use light or sources of light to pass the information. There are
smart phones, medical domains, automated systems, and so on. How­ countless applications for light sources such as LEDs (Weismann et al.,
ever, using these devices results in higher power dissipation and slower 2009) and laser diodes (Chassagneux et al., 2009) etc. These devices are
speed as a result optical devices have been introduced to overcome the relatively cheap, extremely light weight and compact, and quite durable
present modern situations that are to have less power consumption and with a long usable life span. Besides, these solid-state sources generate
high speed. This is due to the fact that optical devices transmit infor­ less heat and require less power compared with more traditional light
mation through light, whereas electronic devices transfer data through sources. Due to their significant energy and cost savings, LEDs are being
electrical signals and the speed of electrical signals is 3.2 m/s, whereas used widely as a replacement source.

* Corresponding author.
E-mail address: drsandipece@gpcet.ac.in (S. Swarnakar).

https://doi.org/10.1016/j.rio.2023.100439
Received 7 February 2023; Received in revised form 29 March 2023; Accepted 30 April 2023
Available online 2 May 2023
2666-9501/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
S. Swarnakar et al. Results in Optics 12 (2023) 100439

The design of optical devices (Priya et al., 2021) can be accomplished ratio, insertion loss, response time, transmission efficiency, and bit rate,
through a variety of means such as photonics (Rao et al., 2021), plas­ were calculated; as a result of these calculations, it is possible to
monics (Abdulnabi and Abbas, 2019), Mach-Zehender interferometer determine how the design will be communicated and Section 4 covers
(Swarnakar et al., 2021), silicon optical amplifier (Oliveira et al., 2018), conclusion.
silicon based ultrafast nonlinear interferometer etc., although these
methods have the drawback that the spreading of light is less when 2. Design procedure of binary subtractor
compared to the photonics method. These approaches use basic building
blocks to create a device such as logic gates, combinational circuits, and In this study, a binary subtractor is developed using the photonic
sequential circuits. A new area of study has emerged as a direct result of crystal (PhC) method, which is often used to minimize the size of optical
the dramatic rise in interest in all-optical signal processing methodolo­ devices. The design of a binary subtractor based on two T-shaped 2-D
gies for use in telecommunications networks. The study of photonic Photonic crystals has been accomplished. Air serves as the background
crystals, which enables the information to be transmitted through pho­ of a square lattice of photonic crystal, which is composed of 14 × 10 rods
tons at a rapid rate, is largely responsible for making all of this possible. of silicon dielectric material. The proposed design utilizes a wafer size of
Due to the compact nature of all-optical devices, photonic crystals are on 8.4 μm × 6 μm. A (14 × 10) array embedded in the XZ plane is the
par with semiconductor technology, with the exception of diffraction. intended form for this structure’s physical manifestation. It allows one
Photonic Crystals also known as PhCs are nanostructures that have a to determine how light is transmitted with varying intensities, such as
periodic arrangement of elements that all have different refractive low and high dielectric substance. The dielectric substance used in this
indices (RIs). In addition, by utilizing a photonics approach, it is possible design is silicon. The permittivity of silicon rod is 11.56, so the refrac­
√̅̅̅̅
to minimize the overall size of the gadget while simultaneously lowering tive index (RI) of the silicon rod is 3.4 (n = εr ) and the radius of each
its overall power consumption. Devices based on PhC can be used to photonic crystal is 0.2 × a; where ‘a’ is the lattice constant which is the
limit the effects of diffraction. To achieve the desired output from ma­ distance between two dielectric rods. The value of lattice constant ‘a’ is
terials of varying refractive index values, PhC devices use Si (Safinezhad 0.6 μm, then the radius of each rod is 0.12 μm. In order to get the desired
et al., 2021), Ge (Hussein et al., 2018), and GaAs (Geerthana et al., output, the design is constructed as shown in Fig. 1b, and it is observed
2022). The properties of each material have an impact on the charac­ that the optimum transmitted optical intensity is attained at the output
teristics of optical logic gates, meaning that they contribute to the port. The designed structure has the 10 rows and 14 columns of silicon
regulation of the flow of light (Liu et al., 2011). A pathway is constructed rods and here the input A is taken at left side, input B is taken at bottom
in this region for the light, which is for the photons to traverse, with the of the right side, R1 is taken left side bottom, at top of the right side is
assistance of line flaws or point defects, which is referred to as channels taken the R2 input, top left side is taken the difference output and right
(Mohebzadeh-Bahabady and Olayee, 2017). The transmission of light side middle is taken the borrow output these things are illustrated in
along the targeted path is facilitated as a result of this. According to its Fig. 1(b).
property, the photonic band gap, also known as PBG, serves to ensure By combining of XOR gate and AND gate with one input compliment
that light cannot flow through it within the wavelength range of the yields the binary subtractor outputs in digital IC. A binary subtractor is
spectrum. In order to operate the transmission of light without affecting combinational circuit which computes the subtractions of two inputs as
structure’s normal behavior, periodic variations in the dielectric con­ shown in the above Fig. 1 (a) these input A and input B computed and
stant are introduced. Under the aforementioned conditions, this study getting the difference and borrow of desired outputs for this design
can be used to suggest and develop sequential circuits and combina­ illustrated in Fig. 1 (b). For the input logic ‘00’ has getting the outputs of
tional circuits. Existing logic gates such as AND (Parandin and Reza logic ‘0’ for both outputs, for the input logic ‘01’ has getting the outputs
Malmir, 2020; Veisi et al., 2021), OR (Derdour et al., 2021; Kotb et al., of logic ‘1’ for both outputs, for the logic input ‘10’ has getting the
2019), NOT (Kumar and Sen, 2020; Kumar and Medhekar, 2019), XOR outputs of logic ‘1’ for difference and logic ‘0’ for borrow and for the
(Kordi et al., 2020; Muthu et al., 2020), XNOR (Asghari et al., 2019; logic ‘11’ has getting the output of logic ‘0’ for both outputs.
Tavakoli et al., 2019), and other gates can used to design combinational A photonic band gap (PBG) (Yablonovitch, 1993; Askarian, 2021) in
(Choudhary et al., 2021) and sequential (Moniem, 2015; Abbasi et al., terms of photon energy distinguishes photonic crystals, which are op­
2012) circuits. For example, combining an XOR gate with an AND gate tical nanostructures with a regularly varying refractive index. Light with
yields half-adder (Swarnakar et al., 2018) outputs, and combining an wavelengths that lie in the photonic band gap cannot exist in the crystal,
XOR gate with one input complement of an AND gate yields half- giving rise to completely different optical phenomena than in free
subtractor (Parandin et al., 2017) outputs. Some digital circuits are space. PBG for the design is shown in Fig. 2 which is simulated using
also designed with different combinations like encoders (Seif-Dargahi, PWE method. PBG range of the design is given by: 0.47879 ≤ λa ≤
2018), decoders (Parandin et al., 2018), and some sequential circuits 0.700731 gap is given by 0.221941. From this range of values, a/λ ratio
like flip-flops (Rao et al., 2020; Zamanian-Dehkordi et al., 2018), reg­ is obtained, here ‘a’ is lattice constant and ‘λ’ is wavelength of contin­
isters, etc. uous wave source. The band gap means that the light could not propa­
At the beginning of this photonic revolution, photonic devices have gate through this specific spectral range. In terms of wavelength, light
become all-pervading in everyday life but often go unnoticed. In high- cannot pass through 0.85 µm to 1.25 µm.
speed data communication, advanced sensing in imaging photonics
presents a growing opportunity for designing and manufacturing de­ 3. Results and performance analysis
vices. Photonic technologies assure the size and speed improvements
with reduced power consumption for data transmission and ultrasensi­ A simulation is a model that duplicates the functioning of a current or
tive sensing capabilities in multiple domains. In this paper, the binary planned system, giving evidence for decision-making by being able to
subtractor circuit is designed using PhC methods. For this, the inputs are test multiple scenarios or process improvements. For a more immersive
A and B, and difference and borrow are the outputs. In general, this work experience, this may be combined with virtual reality technology. A
regarding a binary subtractor has covered section 1 about the develop­ binary subtractor simulation requires four inputs for its Difference and
ment of systems and optical communication using a photonic crystal Borrow outputs: A, B, and two additional inputs. The two additional
structure. In Section 2, the design parameters and structure are dis­ inputs are used to get the desired outputs of the design through
cussed. In Section 3, several computations are investigated, and per­ constructive and destructive interferences. These simulated outputs are
formance analysis is discussed. In this work, it was determined that the compared to the truth table which is shown Table 1 along with binary
constructed device is very small which results in the reduction of size of subtractor output intensities. The binary subtractor is a combination of
the integrated system. In addition, several parameters, such as contrast

2
S. Swarnakar et al. Results in Optics 12 (2023) 100439

Fig. 1. (a) Logical circuit to compute half-subtractor outputs (b) blueprint 2-D structure of binary subtractor.

Fig. 2. Photonic band gap structure in TE mode for designing of binary subtractor structure.

Table 1
Output intensities for refractive index 3.4.
Logic inputs Phase angles Outputs Output intensities

A B R1 R2 A B R1 R2 Diff Bo Diff Bo

0 0 1 1 0 0 0 0 0.09 0.19
◦ ◦
− −
0 1 1 1 180 0 180 1 1 0.46 0.53
◦ ◦ ◦
−
1 0 1 1 180 0 0 1 0 0.66 0.03
◦ ◦ ◦
−
1 1 1 1 0 0 0 180 0 0 0.12 0.01
◦ ◦ ◦ ◦

* Diff: Difference, Bo: Borrow.

XOR and AND gate, for the XOR gate the same input combination can get even integral and phase angle difference is 180 then it gives destructive
◦

the logic low output and for different combination input it produces interference. When the signals collision has odd integral as path differ­
logic high output. For getting the borrow output, a compliment of input ence and phase angle difference is 180 then it is constructive interfer­
◦

A and the AND of inputs A and B is done so only logic ‘01’ can get the ence. When the signals collision has odd integral as path difference and
logic high output and remaining combinations can get the logic low phase angle difference is 0 then it is destructive interference. When two
◦

output. When the light is applied at the input then it is considered as signals superimpose with path difference ‘0’ and phase angle difference
logic ‘1’ and when light is not applied at input then it is considered as is 0 then it is constructive interference. When two signals superimpose
◦

logic ‘0’. The power detected at output port is considered as logic ‘1’ with path difference ‘0’ and phase angle difference is 180 then it is
◦

when it has ≥ 0.4P0ut and if detected power is ≤ 0.3P0ut considered as destructive interference.
logic ‘0’. Case 1: A = B = 0, R1 = R2 = 1
Here, inputs A, B are treated as logic ‘0’, whereas inputs R1 and R2
3.1. Working principal are treated as logic ‘1’ and have a phase angle of 0 . These additional
◦

input from R1 is travelled with 5a path difference to junction J1 and here

Beam interference principle is the one to know how the signal can the more light is reflected towards junction J2 , due to that the difference
make constructive are destructive. According to the wave optic theory, output gets weak signal and input R2 is travelled with ‘4a’ path distance
when two signals superimpose with path difference as an even integral to junction J2 which is reflected towards the junction J1 , due to that weak
and phase angle difference is 0 then it produces constructive interfer­ signal is passing towards borrow output. Since the extra inputs with light
◦

ence. Similarly when two signals superimpose with path difference as an have weak intensity, this is interpreted as logic ‘0’ and these simulated

3
S. Swarnakar et al. Results in Optics 12 (2023) 100439

outputs are shown at Fig. 3(a). The figure shows that no light intensity interference thus there is an absence of light at output borrow. These
passing at the difference and borrow. The outputs obtained by FDTD simulated outputs are shown at Fig. 3(c) as high light intensity passing at
method are also verified with the MATLAB outputs which are shown in the difference and no light intensity is passing at borrow. The outputs of
Fig. 6 (output is getting logic ‘0’ with combination of input A = 0 and FDTD are also verified with the MATLAB outputs which are shown in
input B = 0). Fig. 6 (difference output is getting logic ‘1’ and borrow output getting
Case 2: A = 0, B = 1, R1 ¼ R2 = 1 logic ‘0’ with combination of input A = 1 and input B = 0).
In this instance, input A is treated as logic ‘0’ input B is treated as Case 4: A = B = 1, R1 ¼ R2 = 1
logic ‘1’ with a phase angle of 180 , and further inputs R1 & R2 are In this case, input A and input B are considered as logic ‘1’ with 0
◦ ◦

treated as logic ‘1’ with phase angles of 0 and 180 respectively. The and 0 phase angles and additional inputs R1 & R2 taken logic ‘1’ with
◦ ◦ ◦

wave is primarily coming from R1 to junction J1 , where the path dis­ 0 and 180 phase angles respectively. Here light is inserted from input A
◦ ◦

tance is ‘5a’, then it will travel up to junction J2 with a path distance of and R1 these two waves are meet at the junction J 1 here the path dif­
‘10a’ where the input from B and input R1 have a path difference of ference is 1 a (5a − 4a) and phase angle is 0 due to the odd integral and
◦

a (5a − 4a) which is an odd integral and phase angle difference of 0 that 0 phase angle difference it is determined as destructive interference
◦ ◦

produces a destructive interference. Initially from R1 there is signal due to that output side of difference has no light is passing. Where the
which has ‘10a’ of path difference and at J2 it makes an odd integral with input from B and input R2 have a path difference of 1 a (5a − 4a) and
‘9a’ and the phase angle difference is 180 , by this combination that phase angle difference of 0 and an odd integral by this it can get the
◦ ◦

results in constructive interference. As a result of the constructive phase, destructive interference. Due to this there is no output light is present at
light is present on both outputs of difference and borrow whose outputs the output side borrow and these simulated output is shown at Fig. 3(d)
are logic ‘1’ and these simulated output is shown at Fig. 3(b). The figure at the figure shows that no light intensity passing at the difference and
show that light intensity passing at both difference and borrow is logic borrow. The outputs of FDTD is also verified with the MATLAB outputs
high. The outputs obtained by FDTD method are also verified with the which is shown in Fig. 6 (difference & borrow output is getting logic ‘0’
MATLAB outputs which are shown in Fig. 6 (difference & borrow output with combination of input A = 1 and input B = 1).
is getting logic ‘1’ with combination of input A = 0 and input B = 1). The Table 1 shows the half-subtractor truth table, this table is
Case 3: A = 1, B = 0, R1 ¼ R2 = 1 reference for the binary subtractor design outputs. For the different
In this case, input A is considered as logic ‘1’ with 180 as its phase phase angles, the half-subtractor outputs are computed as shown in
◦

angle, input B is taken as logic ‘0’ and additional inputs R1 & R2 are logic Table 1. The illustrated angles are taken to get the perfect outputs of the
‘1’ with 0 phase angle each respectively. Input light from input A and R1 design and also included the output intensities for the different combi­
◦

will collide at junction J1 and here the path difference is a (5a − 4a) which nation of inputs.
is an odd integral and 180 as phase difference which produces a
◦

constructive interference. Due to that the output is getting high light 3.2. Exploring the different lattice constant computations
intensity at difference output which is logic ‘1’. This constructive wave is
travelled to junction J2 from with junction J1 with a path length of ‘5a’ By adjusting the lattice constant values contrast ratios are calculated
and the path length of additional input R2 to junction J2 is ‘4a’. Here the respectively. Table 2 values are updated with various computations of
path difference is 1a (5a − 4a) which is an odd integral and phase angle lattice constants values. Fig. 4 (a) was made to help visualize the data
difference between them is 180 which produces a destructive presented in Table 2. The optimal lattice constant based on the
◦

Fig. 3. Dispersion of light according to the logical inputs (a) A = B = 0 (b) A = 0, B = 1 (c) A = 1, B = 0 (d) A = B = 1.

4
S. Swarnakar et al. Results in Optics 12 (2023) 100439

Table 2
Different lattice constant with contrast ratio.
Logic inputs Different lattice constants

0.5 0.55 0.6 0.65 0.7

A B Diff Bo Diff Bo Diff Bo Diff Bo Diff Bo

0 0 0.14 0.04 0.32 0.05 0.09 0.19 0.08 0.06 0.18 0.35
0 1 0.01 0.11 0.22 0.108 0.46 0.53 0.45 0.831 1.08 0.37
1 0 0.25 0.10 0.67 0.11 0.66 0.03 0.58 0.30 0.65 0.69
1 1 0.14 0.18 0.51 0.16 0.12 0.01 0.02 0.87 0.10 0.27
Contrast ratio (dB) 13.97 6.53 4.836 5.05 5.185 17.78 14.62 11.61 10.33 4.07

* Diff: Difference, Bo: Borrow.

Note (Table 2): For all the different lattice constants the refractive index and rod radii are same.

Fig. 4. Graphs for different computations of (a) Lattice constant, (b) Refractive index, (c) Rod radii.

calculated values is 0.6 has the better contrast ratio. For 0.5 and 0.55 less intensity light for the output logic ‘1’. For the 0.3a case dispersion of
cases, dispersion of light is not perfect as a result output logic ‘1’ is light intensity is very low for all input combinations so that for these
obtained as logic ‘0’ and for 0.65, 0.7 cases dispersion of light is more for cases cannot suit for the binary subtractor desired outputs.
output logic ‘0’, for these cases the outputs are not perfectly suited for Fig. 4 shows the performance analysis of the device through contrast
binary subtractor output. ratio. This is done by taking different values of lattice constants ranging
from 0.5 to 0.7 and in these variations the best optimal value is
3.3. Exploring the different refractive indexes computations concluded as 0.6, for different values of refractive indexes ranging from
3.2 to 3.45 and in these variations the best optimal value is considered as
By using different Refractive indexes values contrast ratio is calcu­ 3.4, and different values of rod radii ranges from 0.16a to 0.3a and in
lated. For the various computations of refractive indexes values are these variations best optimal value considered as 0.2a. Optimal values of
updated in Table 3 and for the best viewing experience created Fig. 4 (b). refractive index, lattice constant, and silicon rod radius are used to
Best Refractive index of analysis is 3.4. For 3.2, 3.3, 3.35, and 3.45 calculate contrast ratio, transmission efficiency, and insertion loss,
values, more dispersion of light at output logic ‘0’ is observed which which are then used to assess the device’s efficiency.
does not fit for the binary subtractor outputs. The best optimal refractive The quantity of light that is dissipated as the signal travels from the
index is 3.4 in which dispersion of light is good for output logic ‘0’ sending end of the link to the receiving end of the link is called insertion
combination and it has getting the best value for logic ‘1’. loss (Swarnakar et al., 2023) and it is measured in decibels (dB).
Insertion loss = 10log10 (Pout |ON/(PA + PB + PR1 + PR2 ) ) (1)
3.4. Exploring the different rod radii computations
The equation (1) shows the formula to calculate insertion loss where
Pout |ON is considered as highest output intensity and PA +PB +PR1 +PR2 is
By using different rod radii values contrast ratio is calculated. For
the sum of all input intensities which can be obtained by placing
various rad radii computed values are updated in Table 4 and for the
observation lines at inputs to get the input intensities. The calculated
best viewing experience created Fig. 4 (c). For the cases of 0.16a, 0.18a
insertion loss of the device is 0.49 dB. Calculating the contrast ratio
and 0.25a the dispersion of light is more at output logic ‘0’ and getting

Table 3
Contrast ratio with different Refractive indexes.
Logic inputs Different refractive indexes

3.2 3.25 3.3 3.35 3.4 3.45

A B Diff Bo Diff Bo Diff Bo Diff Bo Diff Bo Diff Bo

0 0 0.24 0.18 0.24 0.26 0.23 0.19 0.22 0.21 0.09 0.19 0.18 0.17
0 1 0.41 0.56 0.47 0.63 0.49 0.57 0.45 0.58 0.46 0.53 0.53 0.61
1 0 0.72 0.09 0.72 0.12 0.68 0.04 0.71 0.05 0.66 0.03 10.5 0.02
1 1 0.19 0.05 0.21 0.02 0.22 0.01 0.26 0.01 0.12 0.01 0.25 0.22
Contrast ratio (dB) 5.78 10.4 5.35 14.92 4.9 17.55 4.36 17.63 5.185 17.78 3.65 14.8

* Diff: Difference, Bo: Borrow.

Note (Table 3): For all different Refractive indexes the lattice constant and rod radius is same.

5
S. Swarnakar et al. Results in Optics 12 (2023) 100439

Table 4
Contrast ratio with different rod radii.
Logic inputs Different rod radii

0.16a 0.18a 0.2a 0.25a 0.3a

A B Diff Bo Diff Bo Diff Bo Diff Bo Diff Bo

0 0 0.26 0.19 0.25 0.21 0.09 0.19 0.27 0.13 0.05 0.08
0 1 0.35 0.45 0.45 0.51 0.46 0.53 0.07 0.01 0.02 0.01
1 0 0.63 0.11 0.77 0.14 0.66 0.03 0.45 0.11 0.02 0.03
1 1 0.23 0.10 0.16 0.06 0.12 0.01 0.08 0.38 0.02 0.03
Contrast ratio (dB) 4.37 6.53 6.82 9.29 5.185 17.78 7.501 15.79 3.97 5.74

* Diff: Difference, Bo: Borrow.

Note (Table 4): For all the different Rod radii the Refractive index and Lattice constant is same.

(Swarnakar et al., 2023) allows one to determine how light is trans­ compared with the outputs from each logic input of the binary sub­
mitted with varying intensities, such as low and high. It provides the tractor. When it is compared with logic ‘00’ in Fig. 6 it shows the both
finest possible contrast ratio and it is a good device. outputs logic ‘0’ and Table 1 has shown the low intensity light values
(difference = 0.09, borrow = 0.19), when it is compared for logic ‘01’ in
Contrast ratio (dB) = 10log10 (Pout |ON/Pout |OFF) (2)
Fig. 6 it shows that both outputs are logic ‘1’ and Table 1 shows it has
In equation (2) of contrast ratio formula Pout |ON is highest output produced high intensity light (difference = 0.46, borrow = 0.53), when
intensity whereas Pout |OFF is lowest output intensity. The calculated it is compared for logic ‘10’ in Fig. 6 shows for difference it produced
contrast ratio of the device for difference and borrow is 5.185 dB and logic ‘1’ and borrow produces logic ‘0’ and Table 1 shows it produced
17.78 dB. The transmission efficiency of the bits can also be calculated high intensity light at difference and low intensity light at borrow
which is known as the rate of transmission of bits from input to output. (difference = 0.66, borrow = 0.03), when it is compared for logic ‘11’ in
Fig. 6, it shows that both outputs getting logic ‘0’ and Table 1 shows it
Transmission efficiency = (Pout |ON/(PA + PB + PR1 + PR2 ) ) × 100% (3)
gives low intensity light at both difference and borrow (difference =
The equation (3) shows the formula to calculate transmission effi­ 0.12, borrow = 0.01). These outputs help to verify the outputs produced
ciency where Pout |ON is consider as highest output intensity and by FDTD method.
PA +PB +PR1 +PR2 is the sum of all input intensities which can be ob­ Fig. 7 shows the change in the electric field distribution when
tained by placing observation lines at inputs to get the input intensities. normalized at wavelengths close to 1550 nm. As compared to the input
The calculated transmission efficiency of the device is 89.3 %. power, the output power is lower because losses caused by undesired
Response time includes the time taken to transmit the inquiry, pro­ reflection occur throughout the transmission process. Nevertheless, for
cess it by the device, and transmit the response back to the terminal. error-free detection, the power value of the output when it is in the logic
Response time is frequently used as a measure of the performance of an state ‘1’ should be as close as possible to the maximum input values. On
interactive system. the other hand, the power value of the output when it is in the logic state
‘0’ should ideally be zero. The Fig. 7 shows for every input combination
Bit rate = (1/f s ) (4)
of binary subtractor with different wavelength ranges from 1.45-1.65.
Fig. 7(a) shows the logic ‘00’, Fig. 7(b) shows the logic ‘01’, Fig. 7(c)
where f s given by femtoseconds which is taken from the response time
shows the logic ‘10’ and Fig. 7(d) shows the logic ‘11’ for different
curve shown in the Fig. 5 and for binary subtractor response time is
wavelengths. The three main wavelengths used for fiber optic trans­
calculated as 20.4 f s .
mission are 850, 1300, and 1550 nm. These wavelengths are used in
The difference output port stabilization response time is depicted in
fiber optics because they have the lowest attenuation of the fiber. The
Fig. 5 and it has the 20.4 f s response time. A bit rate of 49.01 Tbps ob­
length of a wave has a direct relationship with its attenuation rate; with
tained from 20.4 f s response time as bitrate is equals to inverse of
longer the wave, lesser the attenuation. 1550 nm is the ideal wavelength
response time. Combination of A = 1 and B = 0 is maximum response
of choice due to its eye safe nature at high power. These high-power
time that is trigger at 20.4 f s , Analyzing the optical output power (as
lasers give users higher fidelity, longer distances and more accurate
shown in Fig. 5) reveals that the output becomes logic ‘1’ at 20.4 f s . At
results than any other option.
20.4 f s , the optical output power exceeds 50% of the normalized input
Table 5 shows the comparison of half-subtractor of previous papers
power. Logic ‘1’ is asserted on the power output and also for combina­
which are collected from online resources. In this paper work, discussed
tion of A = 0 and B = 1 it will trigger at 22.8 f s and its bit rate is 43.8
design is not a new method but the design structure is new one and other
Tbps for borrow.
new parameters are calculated for the design such as contrast ratio,
The Fig. 6 shows the half-subtractor timing waveforms that are
insertion loss, transmission efficiency, bit rate. From ref. (Askarian et al.,
computed with code in MATLAB software. The timing waveforms are
2020; Askarian, 2020; Askarian, 2021; Moradi, 2019; Jiang et al., 2015;
Askarian et al., 2019; Mostafa et al., 2019; Askarian, 2021) it can be
concluded that the size of the devices of compared designs are higher
than this paper work, thus it can be said that it is ultra-compact device,
response time is calculated for this device is 0.0204 Ps and it shows
lower response time than other works, speed of this device is calculated
as 49.01 Tbps as compared to the other work it has high speed.

4. Conclusion

In this article, a novel concept for half-subtractor that makes by using

of two T-shaped waveguides. FDTD and PWE are utilized in order to
stimulate the model. The design is stimulated by utilizing a variety of
various values of lattice constant, refractive index, and silicon rod radii.
Fig. 5. Response time curve for binary subtractor.

6
S. Swarnakar et al. Results in Optics 12 (2023) 100439

Fig. 6. Half-subtractor timing waveforms at MATLAB software.

Fig. 7. The distribution of the electric field at wavelengths close to 1550 nm for the various input combinations of the binary subtractor: (a) A = B = 0; (b) A = 0, B
= 1; (c) A = 1, B = 0; (d) A = B = 1.

This is done in order to obtain values that are optimum for the effec­ Ethical approval
tiveness of the design. It uses the standard optical communication
wavelength of 1.55 μm. The calculated contrast ratio of this design for Not required.
difference and borrow outputs is 5.185 dB and 17.78 dB, a transmission
efficiency of 89.3 %, a response time of 0.024 Ps, and an insertion loss of Funding
0.49 dB are all obtained after analyzing the design. The layout is ultra-
compact and quick in character as compared to other designs in The authors declare that no funds, grants, or other support were
Table 5. Combinational computing can be accomplished with the use of received during the preparation of this manuscript.
this apparatus.
Availability of data and material

The authors can provide the data on request.

7
S. Swarnakar et al. Results in Optics 12 (2023) 100439

Table 5
Comparison table with previous models.
Parameters/ref Ref. (Askarian Ref. ( Ref. ( Ref. ( Ref. (Jiang Ref. (Askarian Ref. (Mostafa Ref. ( This
et al., 2020) Askarian, Askarian, Moradi, et al., 2015) et al., 2019) et al., 2019) Askarian, work
2020) 2021) 2019) 2021)

Mesh size ( μm2 ) 924.6 273.4 627.3 n.r.a 73.5 77 128.52 144.5 50.4
Rod radius (nm) 122 122 120 122 187.5 117 120 117.2 120
Lattice constant (nm) 610 610 585 607 520.8 585 600 586 600
Wavelength (nm) 1550 1550 1551.5 1550 n.r.a 1550 1540 1550 1550
Insertion loss (dB) n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a 0.49
Refractive index 3.46 3.4 3.46 3.46 n.r.a n.r.a 3.4 3.46 3.4
Transmission efficiency n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a 89.3%
Contrast Borrow 7.78 15.05 14.96 6.9 n.r.a 12 3.27 6.56 17.78
ratio ( dB) Difference 6.81 18.80 10.31 8.2 n.r.a 8 5.185

Response time (Ps) 1.5 0.5 0.8 2 n.r.a 1 0.134 0.1 0.0204
Bit rate (Tbps) n.r.a n.r.a n.r.a n.r.a n.r.a n.r.a 7.712 n.r.a 49.01

*n.r.a: not reported.

Declaration of Competing Interest Hussein, H.M., Ali, T.A., Rafat, N.H., 2018. New designs of a complete set of photonic
crystals logic gates. Opt. Commun. 411, 175–181. https://doi.org/10.1016/j.
optcom.2017.11.043.
The authors declare that they have no known competing financial Jiang, Y.C., Liu, S.B., Zhang, H.F., Kong, X.K., 2015. Design of ultra-compact all optical
interests or personal relationships that could have appeared to influence half subtracter based on self-collimation in the two-dimensional photonic crystals.
the work reported in this paper. Opt. Commun. 356, 325–329. https://doi.org/10.1016/j.optcom.2015.07.061.
Jin, C.Y., Wada, O., 2014. Photonic switching devices based on semiconductor nano-
structures. J. Phys. D: Appl. Phys. 47, 1–17. https://doi.org/10.1088/0022-3727/
Data availability 47/13/133001.
Kordi, S.E., Yousefi, R., Ghoreishi, S.S., Adrang, H., 2020. All-optical OR, NOT and XOR
gates based on linear photonic crystal with high port-to-port isolation. Appl. Phys. B
Data will be made available on request. 126, 1–9. https://doi.org/10.1007/s00340-020-07524-2.
Kotb, A., Zoiros, K.E., Guo, C., 2019. Ultrafast performance of all-optical AND and OR
References logic operations at 160 Gb/s using photonic crystal semiconductor optical amplifier
Opt. Laser Technol. 119, 1–10. https://doi.org/10.1016/j.optlastec.2019.105611.
Kumar, A., Medhekar, S., 2019. All optical NOT and NOR gates using interference in the
Abbasi, A., Noshad, M., Ranjbar, R., Kheradmand, R., 2012. Ultra compact and fast All
structures based on 2D linear photonic crystal ring resonator. Optik 179, 237–243.
Optical Flip Flop design in photonic crystal platform. Opt. Commun. 285,
https://doi.org/10.1016/j.ijleo.2018.10.188.
5073–5078. https://doi.org/10.1016/j.optcom.2012.06.095.
Kumar, S., Sen, M., 2020. Integrable all-optical NOT gate using nonlinear photonic
Abdullin, E.N., Morozo, A.V., 2013. Application of rod return-current conductors for
crystal MZI for photonic integrated circuit. JOSA B 37, 359–369. https://doi.org/
decreasing the magnetic field of the vacuum diode. Instrum. Exp. Tech. 56, 420–427.
10.1364/JOSAB.380351.
https://doi.org/10.1134/S0020441213030135.
Li, Z.Y., Ho, K.M., 2003. Light propagation in semi-infinite photonic crystals and related
Abdulnabi, S.H., Abbas, M.N., 2019. All-optical logic gates based on nanoring insulator-
waveguide structures. Phys. Rev. B 68, 1–15. https://doi.org/10.1103/
metal-insulator plasmonic waveguides at optical communication band.
PhysRevB.68.155101.
J. Nanophoton. 13, 016009 https://doi.org/10.1117/1.JNP.13.016009.
Liu, Y., Qin, F., Meng, Z.M., Zhou, F., Mao, Q.H., Li, Z.Y., 2011. All-optical logic gates
Andalib, P., Granpayeh, N., 2009. All-optical ultra-compact photonic crystal AND gate
based on two-dimensional low-refractive-index nonlinear photonic crystal slabs.
based on nonlinear ring resonators. J. Soc. America B 26, 10–16. https://doi.org/
Opt. exp. 19, 1945–1953. https://doi.org/10.1364/OE.19.001945.
10.1088/1464-4258/11/8/085203.
Mohebzadeh-Bahabady, A., Olayee, S., 2017. All-optical NOT and XOR logic gates using
Asghari, M., Moloudian, G., Hassangholizadeh-Kashtiban, M., 2019. A novel proposal for
photonic crystal nano-resonator and based on an interference effect. Inst. Eng. Tech.
all-optical XOR/XNOR gate using a nonlinear photonic crystal based ring resonator.
Optoelectron. 12, 191–195. https://doi.org/10.1049/iet-opt.2017.0174.
Opt. Appl. 49, 1–9. https://doi.org/10.5277/oa190209.
Moniem, T.A., 2015. All-optical SR flip flop using 2-D photonic crystal. Opt. Quant.
Askarian, A., 2020. All optical half subtractor based on threshold switching and beams
Electron. 47, 2843–2851. https://doi.org/10.1007/s11082-015-0173-7.
interference mechanisms. J. Opt. Commun. 1–6. https://doi.org/10.1515/joc-2020-
Moradi, R., 2019. All optical half subtractor using photonic crystal based nonlinear ring
0044.
resonators. Opt. Quant. Electron. 51, 1–9. https://doi.org/10.1007/s11082-019-
Askarian, A., 2021. Design and analysis of all optical half subtractor in 2D photonic
1831-y.
crystal platform. Optik 228, 1–24. https://doi.org/10.1016/j.ijleo.2020.166126.
Mostafa, T.S., Mohammed, N.A., El-Rabaie, E.S.M., 2019. Ultra-high bit rate all- optical
Askarian, A., 2021. All optical half subtractor based on linear photonic crystals and phase
half subtractor based on photonic crystal. National Radio Science Conference IEEE
shift keying technique. J. Opt. Commun. 1–7. https://doi.org/10.1515/joc-2021-
341–349.
0095.
Muthu, K.E., Selvendran, S., Keerthana, V., Murugalakshmi, K., Raja, A.S., 2020. Design
Askarian, A., Akbarizadeh, G., Fartash, M., 2019. A novel proposal for all optical half-
and analysis of a reconfigurable XOR/OR logic gate using 2D photonic crystals with
subtractor based on photonic crystals. Opt. Quant. Elect. 51, 1–9. https://doi.org/
low latency. Opt. Quant. Electron. 52, 1–9. https://doi.org/10.1007/s11082-020-
10.1007/s11082-019-1978-6.
02550-y.
Askarian, A., Akbarizadeh, G., Fartash, M., 2020. An all-optical half subtractor based on
Oliveira, J.M., Sousa, F.B.D., Sousa, F.M.D., Oliviera, J.E.D., Costa, M.B.C., Costa, M.B.C.,
Kerr effect and photonic crystals. Optik 207, 1–16. https://doi.org/10.1016/j.
2018. A new system for all-optical AND logic gate on Semiconductor optical amplifer
ijleo.2020.164424.
based Michelson. Int. Symp. Instrum. Sys. Circuit Transducer 1–6. https://doi.org/
Calhoun, B.H., Cao, Y., Li, X., Mai, K., Pileggi, L.T., Rutenbar, R.A., Shepard, K.L., 2008.
10.1109/INSCIT.2018.8546703.
Digital circuit design challenges and opportunities in the era of nanoscale CMOS.
Parandin, F., Reza Malmir, M., 2020. Reconfigurable all optical half adder and optical
Proc. IEEE 96, 343–365. https://doi.org/10.1109/JPROC.2007.911072.
XOR and AND logic gates based on 2D photonic crystals. Opt. Quant. Electron. 52,
Chassagneux, Y., Colombelli, R., Maineult, W., Barbieri, S., Beere, H.E., Ritchie, D.A.,
1–8. https://doi.org/10.1007/s11082-019-2167-3.
Khanna, S.P., Linfield, E.H., Davies, A.G., 2009. Electrically pumped photonic-crystal
Parandin, F., Malmir, M.R., Naseri, M., 2017. All-optical half-subtractor with low-time
terahertz lasers controlled by boundary conditions. Nature 457, 174–178. https://
delay based on two-dimensional photonic crystals. Superlattices and Microstructures
doi.org/10.1038/nature07636.
109, 437–441. https://doi.org/10.1016/j.spmi.2017.05.030.
Choudhary, K., Singh, A., Singh, A., Chaudhary, D., Kumar, S., 2021. Implementation of
Parandin, F., Karkhanehchi, M.M., Naseri, M., Zahedi, A., 2018. Design of a high bitrate
highly optimized optical all logic gates on a single chip using Ti-diffused lithium-
optical decoder based on photonic crystals. J. Comp. Electron. 17, 830–836. https://
niobate for high-speed processing in combinational circuits. Microelectron. J. 111,
doi.org/10.1007/s10825-018-1147-3.
1–7. https://doi.org/10.1016/j.mejo.2021.105048.
Priya, N.H., Swarnakar, S., Krishna, S.V., Kumar, S., 2021. Design and analysis of a
Derdour, R., Lebbal, M.R., Mouetsi, S., Hocini, A., 2021. A novel connected structure of
photonic crystal-based all-optical 3-input OR gate for high-speed optical processing.
all-optical high speed and ultra-compact photonic crystal OR logic gate. J. Opt.
Opt. Quant. Electron 53, 1–7. https://doi.org/10.1007/s11082-021-03374-0.
Commun. https://doi.org/10.1515/joc-2021-0152.
Rao, D.G., Palacharla, V., Swarnakar, S., Kumar, S., 2020. Design of all-optical D flip-flop
Geerthana, S., Syedakbar, S., Sridarshini, T., Balaji, V.R., Sitharthan, R., Shanmuga
using photonic crystal waveguides for optical computing and networking. Appl. Opt.
Sundar, D., 2022. 2D-PhC based all optical AND, OR and EX-OR logic gates with high
59, 7139–7143. https://doi.org/10.1364/AO.400223.
contrast ratio operating at C band. Laser Phys. 32 (10), 106201.

8
S. Swarnakar et al. Results in Optics 12 (2023) 100439

Rao, D.G.S., Swarnakar, S., Kumar, S., 2021. Design of photonic crystal based compact plasmonic waveguides for optical networks. Opt. Quant Elect. 55, 1–15. https://doi.
all-optical 2 ×1 multiplexer for optical processing devices. Microelectron. J. 112, org/10.1007/s11082-022-04329-9.
1–6. https://doi.org/10.1016/j.mejo.2021.105046. Tavakoli, G., Alipour-Banaei, H., Hassangholizadeh-Kashtiban, M., 2019. A novel
Safinezhad, A., Babaei Ghoushji, H., Shiri, M., Rezaei, M.H., 2021. High-performance proposal for all-optical compact and fast XOR/XNOR gate based on photonic crystal.
and ultrafast configurable all-optical photonic crystal logic gates based on J. Modern Opt. 66, 599–605. https://doi.org/10.1080/09500340.2018.1559947.
interference effects. Opt. Quant. Electron. 53, 1–20. https://doi.org/10.1007/ Veisi, E., Seifouri, M., Olyaee, S., 2021. A novel design of all-optical high speed and ultra-
s11082-021-02856-5. compact photonic crystal AND logic gate based on the Kerr effect. Appl. Phys. B 127,
Schneider, B., Strutt, M.J., 1959. Theory and experiments on shot noise in silicon pn 1–9. https://doi.org/10.1007/s00340-021-07618-5.
junction diodes and transistors. Proc. IRE 47, 546–554. https://doi.org/10.1109/ Weismann, C., Bergenek, K., Linder, N., Schwarz, U.T., 2009. Photonic crystal
JRPROC.1959.287337. LEDs–designing light extraction. Las. Photon. Rev. 3, 262–286. https://doi.org/
Seif-Dargahi, H., 2018. Ultra-fast all-optical encoder using photonic crystal-based ring 10.1002/lpor.200810053.
resonators. Photon. Netw. Commun. 36, 272–277. https://doi.org/10.1007/s11107- Yablonovitch, E., 1993. Photonic band-gap structures. J. Soc. America B 10, 283–295.
018-0779-3. https://doi.org/10.1364/JOSAB.10.000283.
Swarnakar, S., Kumar, S., Sharma, S., 2018. All-optical half-adder circuit based on beam Zamanian-Dehkordi, S.S., Soroosh, M., Akbarizadeh, G., 2018. An ultra-fast all-optical RS
interference principle of photonic crystal. J. Opt. Commun. 39, 13–17. https://doi. flip-flop based on nonlinear photonic crystal structures. Opt. Rev. 25, 523–531.
org/10.1515/joc-2016-0121. https://doi.org/10.1007/s10043-018-0443-2.
Swarnakar, S., Guddati, A., Reddy, S.K., Harijan, R., Kumar, S., 2021. Performance Zhang, W., Wang, Z., Singh, R., Wang, Y., Xie, Y., Su, X., Gao, F., Li, G., Swarnakar, S.,
analysis of optimized plasmonic half-adder circuit using Mach-Zehnder Min, R., Zhang, B., Kumar, S., 2022. Performance analysis of an SMF-/MMF-based
interferometer for high-speed switching applications. Microelectron. J. 111, 1–6. single/double/quadruple tapered optical fiber structure. Appl. Opt. 61, 2140–2146.
https://doi.org/10.1016/j.mejo.2021.105040. https://doi.org/10.1364/AO.453486.
Swarnakar, S., Basha, S.C., Azmathullah, S., Prabhu, N.A., Madhu, G., Kumar, S., 2023. Zhang, Y., Zhang, Y., Li, B., 2007. Optical switches and logic gates based on self-
Improved design of all-optical half-adder and half-subtractor circuits using MIM collimated beams in two-dimensional photonic crystals. Opt. Exp. 15, 9287–9292.
https://doi.org/10.1364/OE.15.009287.