Conventional Biophotonic Sensing Approach for Sensing and Detection of Normal and Infected Samples Containing Different Blood Components

Abstract

The present research work focused on the biosensing capabilities of one-dimensional (1D) defected photonic crystal (PC). This proposed structure is capable of simultaneously discriminating between normal and infected samples containing the platelet, plasma, and hemoglobin components of blood. The proposed biosensor was designed by creating a defect layer D of air in the middle of 1D PC (PQ)^N to form modified 1D PC with a defect as (PQ)^N/2D(PQ)^N/2. The period number (N) of 1D PC (PQ)^N was chosen to be 10. The cavity region of air was used to investigate only one of the five samples, at each time, that are part of this study. The theoretical findings of the proposed work were obtained using the well-known transfer matrix method in addition to MATLAB software. The results were computed corresponding to defect layer thicknesses of 200 and 700 nm under normal incidence to overcome the difficulties associated under oblique incidence corresponding to TE and TM polarized waves. We examined the performance of the proposed design by calculating the sensitivity, figure of merit, and quality factor values of the biophotonic sensor loaded with different samples. It was found that the sensitivity of our design reaches to a maximum of 428.6 nm/RIU corresponding to the defect layer thickness of 700 nm, when the cavity is loaded with an infected blood sample containing plasma only. This study successfully simultaneously examined the normal and infected blood samples containing the platelet, plasma, and hemoglobin components of blood.

1. Introduction

The emergence of many diseases and health-related problems in recent years has led to the development of early detection and sensing mechanisms that enable timely diagnose of these diseases. Among the best recent technologies that have proven their worth in early sensing is the biosensor based on PC structures [1,2,3]. PCs are the periodic nanostructures that can control propagation of photons passing through them, similar to that of the propagation of electrons passing through semiconductors. PCs have photonic band gap (PBG) regions that appear due to the periodic nature of the structure. The incident electromagnetic radiations of frequencies that overlap with the PBG are not able to pass through the structure and hence have zero transmittance of light in the PBG region [4,5,6,7,8,9]. The significant effective control of light passing through PCs due to the existence of the PBG has attracted scientists to develop novel devices that have potential applications in the field of photonic engineering and technology. The direction of modulation of the refractive index of constituent material layers of PC can be in one direction (1D), two directions (2D), or three directions (3D). Keeping the direction of modulation of the refractive index in mind, the PCs can be classified into three categories, namely, 1D, 2D, or 3D [10]. The photonic biosensors composed of 2D multilayer periodic structures are widely used for biosensing applications because they have higher confinement of light with better control over propagating modes [11,12,13]. The fabrication of 2D photonic biosensors is less complicated in comparison to the fabrication of 3D photonic structures. Recently, a nano-size ring resonator design composed of 2D PC was reported by Bahabady et al. This design can be used for the detection of the concentration of glucose in blood, and for determining seawater salinity [14]. By comparison, due to the ease of fabrication, higher sensitivity, and limit of detection, biosensors comprising 1D PCs are more suitable for biosensing applications. Moreover, photonic devices composed of 1D PC have a high degree of accuracy. This has attracted the worldwide attention of researchers to design and develop different 1D photonic devices suitable to be used in diversified applications such as filters [6], switches [7], photonic papers, inkless printing, and smart windows [15,16,17,18,19]. The insertion of an additional defect layer into PC breaks its periodicity, due to which a defect mode, also called the resonant tunneling mode of appreciable transmittance, exists inside the PBG region [20,21]. The optical properties associated with this resonant mode can be easily manipulated by changing the refractive index of the defect layer region. This intriguing property of PC with a defect is now being used for the design of a different kind of biosensor, due to their high demand in the fields of chemical engineering, biomedical sciences, and food technologies. These photonic biosensors provide rapid, accurate, and economic results.

In recent years, photonic biosensors have received a large amount of attention in the analysis of different components of human blood, due to their rapid and accurate results. To carry out these investigations, the biosensors require a very small volume of blood in the sample. The dengue virus is among the tropical diseases that scientists want to detect at an early stage to ensure timely treatment [20,21]. The dengue virus enters into human blood through the bite of Aedes mosquitoes [21,22,23], which causes a high fever. The other symptoms associated with dengue, such as muscle, bone, or joint pain, nausea, pain behind the eyes, and headaches, usually appear after 2 to 7 days [24,25]. The dengue virus may become dangerous and can cause hemorrhagic fever, which usually affects children under 15 years of age. Thus, in this study we proposed a design for sensing and detecting the dengue virus in blood using 1D PC with a defect. The working principle of our design is to detect the change in the position of the defect mode inside the PBG due to the change in the refractive index of the normal and infected blood samples separately containing platelet, plasma, and hemoglobin components. Basically, the change in the refractive index of different blood samples is the cornerstone of various biological and physical variations. In the proposed work, we took three blood samples separately containing platelet, plasma, and hemoglobin of normal and infected people [26]. We used the well-known standard transfer matrix method (TMM) and MATLAB software to simulate the results of our work. Because the details of this method can be easily obtained from many research papers published in international journals of repute by the same research group, in this work, we do not provide the details of the TMM used [27,28,29,30].

The organization of our manuscript is as follows. The introduction was provided in Section 1. The theoretical formulation of the proposed structure is presented in Section 2. The results and discussion are presented in Section 3. Finally, conclusions are given in Section 4.

2. Theoretical Formulation

Herein, we describe the theoretical formation pertaining to the proposed (PQ)^N/2D(PQ)^N2/2 design. Here, the letters P, Q, and D represent the layers of materials Si, ZnO, and air, respectively. The period number is represented by the letter N. The defect layer D of air, which was created for the proposed design, is used to analyze different blood samples under investigation. The whole structure is surrounded by air. Let us suppose that the plane electromagnetic wave impinges on the structure normally from air (Figure 1).

The transfer matrix [11,12,13,14,15,16,17,18,19,20,21] that describes the interaction of the incident plane electromagnetic wave with the structure under normal incidence is given as:

$$L=\left(\begin{array}{cc}{L}_{11}& L_{12}\\ L_{21}& L_{22}\end{array}\right)={\left(l_1{l}_2\right)}^{N/2}\left(l_3\right){\left(l_1{l}_2\right)}^{N/2}$$

(1)

where $L_{11}, L_{12}, L_{21}, and L_{22}$ are the elements of transfer matrix. N is the period number of structure. The characteristic matrices representing the layers of materials P, Q, and D are denoted by l₁, l₂, and l₃ respectively, and are defined as:

$$l_z=\left(\begin{array}{cc}cos{\gamma }_z& -i/q_{z}sin{\gamma }_z\\ -i{q}_{z}sin{\gamma }_z& -i/q_{z}sin{\gamma }_z\end{array}\right)$$

(2)

where the index $z$ = P, Q, D denotes different material layers of the structure. For the TE polarized wave, the value of q_z under normal incidence is given by $q_z=n_z$. Here ${\gamma }_z$ is the phase difference at each layer, which is defined as:

$${\gamma }_z=\frac{2\pi n_z{d}_z}{\lambda }$$

(3)

In this equation, n_z, d_z, and α_z are the refractive index, thickness, and ray angle inside each layer, respectively.

The transmission coefficient t of the proposed structure is given by:

$$t=\frac{2q_0}{\left(L_{11}+L_{12}{q}_S\right)q_0+\left(L_{21}+L_{22}{q}_S\right)}$$

(4)

where the values of $q_0=n_0$ and $q_S=n_S$ are associated with the incident and exit media of the structure, respectively, corresponding to the TE polarized wave. The transmittance (T) of the proposed biosensing structure is given by:

$$T=\frac{q_s}{q_0}{\left|t\right|}^2$$

(5)

3. Results and Discussion

The present research work is based on the study of propagation of light through 1D PC, whose periodicity is broken by introducing a defect layer of water having a thickness d_D = 200 nm at the middle of 1D PC (PQ)^N to form a defect PC (PQ)^N/2D(PQ)^N/2, as shown in Figure 1. Here, the letters P and Q represent Si and ZnO material layers, respectively. The thickness of the layers P and Q are represented by d_P and d_Q, respectively, such that d_P = 60 nm and d_Q = 40 nm. The refractive indices of layers P and Q are denoted by n_P and n_Q, respectively. Si and ZnO materials were selected in our design due to the ease of fabrication of the photonic designs made up of these materials [31]. We are certain that the proposed biophotonic design can be easily fabricated by depositing alternate layers of Si and ZnO materials on glass substrate using the presently available thin film disposition techniques. Methods such as spin coating, magnetron sputtering, sol-gel, and chemical vapor deposition can be used for fabrication of the proposed design. Due to its flexibility, the spin coating fabrication technique, which can be applied to a wide range of materials and nanoparticles, is superior to other approaches. To create an air cavity of thickness d_D, electron beam lithography or dry chemical etching can be used by applying strong chemicals for the removal of the preferred layer from the fabricated multilayer structure. These kinds of 1D photonic structure having a defect have been previously successfully fabricated [32,33,34], which motivated us to design the proposed biophotonic sensor.

The numeric values of n_P and n_Q were taken to be 3.3 and 2, respectively [35,36,37]. The refractive index of the defect layer of water was chosen to be n_D = 1.333. The period number N of the structure was fixed to 10 to obtain a wider PBG, which extended from 530 to 680 nm for a width of 150 nm, as shown in Figure 2. The figure also shows the resonant tunneling peak having transmittance of 98%, and centered at wavelength 640 nm, inside the PBG due to the break in periodicity of 1D PC, which is highly dependent on the thickness and refractive index of the defect layer region. All the investigations of this research were carried out under normal incidence conditions, thereby avoiding the difficulties associated with oblique incidence due to the dependence of TE and TM polarization cases.

Next, we replaced the water sample from the cavity and poured the various normal and infected blood samples separately containing platelet, plasma, and hemoglobin components into the cavity. In the present work, we simultaneously investigated a total of six samples separately containing normal and infected platelet, plasma, and hemoglobin components of blood. The refractive index of the blood samples varied from 1.337 to 1.400, corresponding to infected plasma and hemoglobin, respectively. Based on the refractive index variation in the samples under investigation, we selected the structural parameters to obtain a wider PBG, in addition to using the position of defect mode centered at the middle of this PBG. This was because, due to the change in the refractive index of the blood sample, the position of the defect mode may modulate toward higher or lower PBG edges. If the position of the defect mode is in the middle, depending on the range of the sample under consideration, its modulation would be easily detectable. Moreover, we would be able to accommodate a large number of blood samples.

The refractive index of the blood sample containing platelet and plasma, corresponding to both normal and infected samples are published in different papers [38,39,40,41].

Figure 3, Figure 4 and Figure 5 show the transmission plots of the proposed structure loaded with normal and infected blood samples of the platelet, plasma, and hemoglobin components, respectively. Figure 3, Figure 4 and Figure 5 show that more than 95% of the two distinguishable and distinct defect modes of each transmission are inside the PBG extending from 620 to 680 nm, corresponding to samples containing platelet, plasma, and hemoglobin components of blood, respectively. The two defect modes in each figure are distinguished by blue and red colors, which correspond to the normal and infected blood samples containing platelet, plasma, and hemoglobin components of blood. It is evident from Figure 3 and Figure 4 that the defect modes associated with blood samples of a healthy person containing platelet and plasma components exist at a higher wavelength, in comparison to the defect modes associated with infected blood samples containing platelet and plasma components.

Moreover, the separation between defect modes is corresponding to normal and infected samples separately containing platelet and plasma components decreases when changing from the platelet sample to the plasma sample, as evident from Figure 3 and Figure 4, respectively. This decrease in separation is due to the fall in the refractive index values of normal and infected samples when the sample was changed from platelet to plasma. The change in sample also results in the shifting in the defect modes corresponding to normal and infected samples containing platelet and plasma components towards a lower wavelength. This movement in the defect modes is due to the decrease in the refractive index of the blood sample containing platelet and plasma, corresponding to both normal and infected samples [38,39,40,41]. An additional finding is that the defect modes corresponding to normal and infected samples separately containing the platelet and plasma components are found at higher and lower wavelengths, respectively, inside the PBG. The reason for this is the denser and rarer refractive index values of normal and infected samples, respectively, separately containing the platelet and plasma components, as evident in different papers [38,39,40,41]. However, the findings of the sample containing the hemoglobin component are in contrast to those of the samples containing the platelet and plasma components, as shown in Figure 5. The figure shows that the defect modes corresponding to normal and infected samples containing the hemoglobin component of blood inside the PBG are found toward lower and higher wavelengths, respectively. A common observation from Figure 3, Figure 4 and Figure 5 is that the intensity of the defect mode is marginally higher than that of normal or infected samples whose refractive index is greater.

The working efficiency and competence of the proposed biophotonic sensor, which is capable of sensing normal and infected samples containing platelet, plasma, and hemoglobin components in blood, was investigated by evaluating the sensitivity (S), figure-of-merit (FoM), and quality factor (Q_f) values of the proposed design [11,12,13,14,15]. Prima facie, these are the basic parameters to be calculated for the evaluation of detection and sensing capabilities of any efficient biophotonic sensor design. The minute change in the refractive index of the normal and infected samples containing particular blood components, $\left(n_D^{normal}-n_D^{infected}\right)$, results in the corresponding shift in the position of the defect mode, $\left({\lambda }_D^{normal}-{\lambda }_D^{infected}\right)$, inside the PBG. This is defined as:

$$S=\frac{\left({\lambda }_D^{normal}-{\lambda }_D^{infected}\right)}{\left(n_D^{normal}-n_D^{infected}\right)}$$

(6)

The capability of the proposed biophotonic sensor to determine very minute positional changes in the defect mode inside the PBG is directly proportional to the sensitivity of the design and inversely proportional to the full width half maximum (FWHM) of the defect mode associated with the sample under investigation. The FoM is defined as:

$$FoM=\frac{S}{FWHM}$$

(7)

The ability of the biophotonic sensor that determines the narrow band width is defined in terms of its quality factor Q_f as:

$$Q_f=\frac{{\lambda }_D}{FWHM}$$

(8)

The S, FoM, and Q_f values of the proposed design loaded with normal and infected samples containing platelet, plasma, and hemoglobin blood components were determined using Equations (6)–(8), respectively, in order to evaluate the performance of the proposed biosensing structure. The numeric values of S, FoM, and Q_f, corresponding to the normal and infected samples containing the platelet, plasma, and hemoglobin blood components listed different papers [38,39,40,41], are listed in

Table 1. The calculated numeric values of S, FoM, and Q_f of the proposed 1D biophotonic sensor structure loaded with normal and infected samples containing platelet, plasma, and hemoglobin blood components.

Blood Component	Classification	nD (RIU)	λD	FWHM (nm)	S (nm/RIU)	FoM	Qf
Platelet	Normal	1.390	650.6	0.075	166.6	2221.3	8674.6
	Infected	1.357	645.2	0.050	170.0	3400.0	12,904.0
Plasma	Normal	1.350	644.0	0.055	157.1	3090.9	11,709.1
	Infected	1.337	641.7	0.050	170.0	3142.0	12,834.0
Hemoglobin	Normal	1.360	645.7	0.055	166.0	3018.2	11,740.0
	Infected	1.400	652.2	0.075	165.7	2209.3	8696.0

.

Table 1 shows the sensitivity of the proposed biophotonic sensor loaded with normal and infected blood samples separately containing platelet, plasma, and hemoglobin components. The sensitivity of the structure under the influence of various samples varies between a maximum of 170.0 nm/RIU and a minimum of 157.1 nm/RIU. The average values of FoM and Q_f are 2846.95 and 11,092.95, respectively. The average values of FoM and Q_f are more than enough to make our biosensor suitable for detecting very minute changes in the refractive index of normal and infected samples separately containing platelet, plasma, and hemoglobin blood components.

Next, we plotted Figure 6, which shows the central wavelength of the defect mode as a function of the refractive index of normal and infected samples containing platelet, plasma, and hemoglobin blood components. It can be observed that the central wavelength of defect modes (λ_D) corresponding to different samples increases linearly with the refractive index of various samples (n_D) containing different blood components. A linear curve was fitted to the simulated data to obtain the following relation:

$${\lambda }_D\left(nm\right)=152.8n_D\left(RIU\right)+437.96\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\left(R^2=0.99185\right)$$

(9)

where RIU represents the refractive index unit. The correlation coefficient between the fitted linear curve and the simulated data is shown by R². The R² value corresponding to the calculated data is 0.99185, which shows that the performance of our design is highly linear. Hence, the results of the proposed biosensor are highly linear. The slope of the fitted curve highlights the sensitivity of the proposed design, which is 152.8 nm/RIU.

Furthermore, we studied the effect of changes in the refractive index of various samples containing different blood components on the sensitivity of the proposed design. For this purposed, we plotted Figure 7, which shows the significance of the refractive index of various samples containing different blood components on the sensitivity. As shown in Figure 7, the sensitivity increases rapidly from 155 to 170 nm/RIU, corresponding to the change in the refractive index of the sample from 1.33 to 1.35. The further increase in the refractive index from 1.35 to 1.37 results in a decrease in the sensitivity from 170 to 166.2 nm/RIU; following a further increase in the refractive index of the sample under investigation from 1.37 to 1.40, the sensitivity of the structure increases slowly and reaches the maximum value of 166.35 nm/RIU at n_D = 1.395. The minimum value is 165.7 nm/RIU at n_D = 1.40.

Finally, we increased the thickness of the defect layer region from d_D = 200 to d_D = 700 nm, and simultaneously loaded the cavity region with normal and infected blood samples separately containing platelet, plasma, and hemoglobin components. Under these circumstances, the performance of the proposed design was evaluated by calculating the sensitivity of the structure loaded with infected samples with respect to normal samples separately containing platelet, plasma, and hemoglobin blood components. The increase in the thickness of the cavity region from 200 to 700 nm results in the movement in the defect mode towards a higher wavelength, in addition to a significant improvement in the sensitivity. These results are similar to the observations reported previously [32,33]. The further increase in d_D beyond 700 nm results in a slight increase in the sensitivity, but also enhances the requirement of the volume of the blood sample. Additionally, the increase in d_D beyond 700 nm also results in the movement in the defect mode towards a higher wavelength, until it leaves the PBG and a new defect mode enters the PBG, as reported by Arafa et al. [30]. Due to the above-mentioned reasons, we chose the optimum value of the defect layer thickness as d_D = 700 nm. The maximum values of the sensitivity of the proposed structure, whose cavity is loaded with normal and infected blood samples separately containing platelet, plasma, and hemoglobin components under normal incidence, are listed in

Table 2. The calculated numeric value of S of the proposed 1D biophotonic sensor structure having a cavity thickness of d_D = 700 nm loaded with normal and infected samples containing platelet, plasma, and hemoglobin blood components.

Blood Component	Classifications	Refractive Index	Senstivity (nm/RIU)
Platelet	Normal	1.390	300.0
	Infected	1.357	329.6
Plasma	Normal	1.350	345.0
	Infected	1.337	428.6
Hemoglobin	Normal	1.360	277.1
	Infected	1.400	292.9

. For all three samples separately containing platelet, plasma, and hemoglobin blood components, the sensitivity of the structure with the infected sample is more than the sensitivity of the structure loaded with the normal sample.

We further plotted the diagram showing the dependence of the sensitivity of the structure loaded with different samples on their refractive index, as shown in Figure 8. It is evident from Figure 8 that the sensitivity of the structure reaches a maximum of 428.6 nm/RIU with the infected plasma sample whose refractive index is the lowest i.e., 1.337; the minimum of 277.1 nm/RIU is achieved with the normal hemoglobin sample under investigation.

Finally,

Table 3. Performance evaluation table showing the comparison between the results of the present work and the previous work of a similar type for the proposed photonic structure (PQ)^N/2D(PQ)^N/2.

Year	S (nm/RIU)	Q-Factor	FoM (RIU)	Sample Type	Reference
2019	48.6–90.9	Not mentioned	Not mentioned	Blood	[38]
2019	25.75–51.49	Not mentioned	Not mentioned	Blood	[39]
2019	0.83	Not mentioned	Not mentioned	Blood	[40]
2021	203.09	1569	Not mentioned	Blood	[41]
2021	71–75	Not mentioned	Not mentioned	Blood	[1]
This work	277.1–428.6	103–104	104	Blood

presents a comparison between the results of this work and the excellent research of distinguished researchers. The aim of this table is to highlight the findings of the proposed work in contrast to the previous research work carried out by various research groups. It can be easily understood from Table 3 that our structure is very sensitive in comparison to earlier reported work due to the larger value of sensitivity, which varies between 277.1 to 428.6 nm per RIU. In addition, the quality factor and FoM values of our design are either higher or of same order of magnitude in comparison to the earlier reported work. All the work presented in Table 3 for comparison is based on photonic blood sensing technology, thus enabling a true comparison to be made.

4. Conclusions

In conclusion, we proposed a conventional approach for simultaneously sensing and detection of normal and infected samples separately containing platelet, plasma, and hemoglobin blood components. The proposed design is composed of 1D PC with a defect. All the investigations pertaining to the work were carried out under normal incidence to overcome the issues related to oblique incidence. The sensitivity of the proposed design was increased significantly by increasing the thickness of the cavity region from 200 to 700 nm. In addition to sensitivity, we also calculated FoM and Q_f values of the proposed design to judge the design’s performance. The FoM merit values of our design were sufficiently high, which signifies that the proposed design is capable of identifying the very minute changes in the position of the defect mode inside the PBG of 1D PC with a defect, due to changes in the sample under investigation depending on the refractive index values of the respective sample. Moreover, the order of the Q_f values of the proposed design varied from 10³ to 10⁴, which is also large. This indicates the ability of our biophotonic sensor to determine the narrow band width of the defect mode corresponding to every sample.