Optical Nano Antennas: State of the Art, Scope and Challenges as a Biosensor Along with Human Exposure to Nano-Toxicology

The concept of optical antennas in physical optics is still evolving. Like the antennas used in the radio frequency (RF) regime, the aspiration of optical antennas is to localize the free propagating radiation energy, and vice versa. For this purpose, optical antennas utilize the distinctive properties of metal nanostructures, which are strong plasmonic coupling elements at the optical regime. The concept of optical antennas is being advanced technologically and they are projected to be substitute devices for detection in the millimeter, infrared, and visible regimes. At present, their potential benefits in light detection, which include polarization dependency, tunability, and quick response times have been successfully demonstrated. Optical antennas also can be seen as directionally responsive elements for point detectors. This review provides an overview of the historical background of the topic, along with the basic concepts and parameters of optical antennas. One of the major parts of this review covers the use of optical antennas in biosensing, presenting biosensing applications with a broad description using different types of data. We have also mentioned the basic challenges in the path of the universal use of optical biosensors, where we have also discussed some legal matters.


Introduction
The optical antennas, which represent unique optical detectors equivalent to radio frequency (RF) antennas, are a novel concept in the field of physical optics [1]. The optical antenna is an helping tool for influencing and regulating radiation in the optical regime. Nowadays, optical antennas are subjected to an increasing amount of technical research. This technology has potential in the enhancement of the efficiency of sensing, light emission, photo-detection, spectroscopy, and heat transfer [1]. Conventionally, optics and photonics are involved in the regulation of optical propagation using fibers, lenses, mirrors, and different diffractive components. In almost all areas, antennas are universal, covering satellite to toys. As optical antennas have numerous prospects, the key benefits of this type of antennas can be précised as follows: Optical antennas: (i) are point detectors which secure a recognition space of almost the square of the wavelength [2].
(ii) combine optical radiation into minute volumes for generating currents in the wire which are identified by a rectifying component of almost 0.02 µm 3 volume. This minute material volume permits one to achieve faster responses. Initial assessments of this response time are about 100 ns for devices without optimization [3]. Conversely, one of the rectifying tools employed in detecting the signal is constructed on the basis of a tunnel effect, which has a response time of approximately 10-14 s, 10-15 s [4]. (iii) are known as polarization-sensitive sensors like the RF versions [2]. (iv) are capable of being tuned to a particular wavelength region. At optical frequencies, the metallic structures have a lossy character and as a result, the resonances are likely to be widened, which possibly limits the tuning ability [5]. (v) are directionally sensible subject to the metallic structure design and the addition of peripheral optical devices [6].
Though the optical antenna has use possibilities in numerous fields, it has a great possibility for use as a biosensor and this review only highlights the biosensing application. This review provides a clear overview of optical biosensors to the reader, a concept that arises from the contact of visible light with free electrons at a metal-dielectric boundary [7].

History of Optical Antennas
The root of the theory of optical antenna can be found in near-field optics [8]. The proposal of using a colloidal gold nanoparticle for optical radiation concentration on a model surface to overcome the restrictions of diffraction in imaging is first made by Synge in 1928 [9]. The concept of using gold nanoparticles as an antenna was first presented in 1985 by Wessel [10] and it was first demonstrated experimentally by using a gold-coated polystyrene particle by Fischer et al. in 1995 [11]. In the succeeding years, sharply pointed optical antennas were used in microscopy and spectroscopy [12][13][14]. Tip-enhanced near-field optical microscopy is the result of these experiments. In early 1968, optical antennas were utilized as whisker diodes in infrared radiation recognition and combination [15][16][17] and as a continuation of these studies, various investigations about infrared antenna structures have been done [18][19][20].
In 1997, after proof of principle experiments, bow-tie type antennas have been suggested as optical probes for the near-field regime [21]. Later investigations presented the fabrication of bow-tie type antennas on tips [22]. After the establishment of the similarity of optical antennas with near-field optical probes [8], tip-on-aperture probe techniques become popular to grow the antenna structures [23,24]. As a result of these advances, many researchers head off to explore various antenna geometries with both experimental and theoretical approaches. As an example, Figure 1 displays several antenna shapes fabricated using different techniques. Nowadays, the use of surface plasmon resonance in optical antennas makes them more efficient for selected frequencies which holds potential for sensing and detection in the field of biology [18,[25][26][27][28][29][30][31][32].

Physical Properties of Optical Antenna
Before entering in depth into the field of optical antennas, we should know their basics. The main parameters for designing optical antennas are:

Local Density of Electromagnetic States (LDOS)
In the discussion of antennas, one of the most significant parameters is impedance. According to circuit theory, impedance is defined as Z = V/I, where I is current and V is voltage. According to this definition, the antenna is connected to the source through a transmission line, but this definition of antenna input impedance needs to be modified due to the feeding of optical antennas by confining light emitters rather than real currents. A practical replacement of this definition comprises the LDOS. This LDOS is the cause of the dipole energy dissipation in a random inconsistent environment. The allowance of a clear relationship of quantum-conventional formalisms is the main benefit of using the LDOS. LDOS is represented by ρ and the total LDOS can be found as [28]: where Tr indicates the trace, ρp is the partial LDOS, ω is the transition frequency, G is the Green function tensor, c is the velocity of light, and r0 is an arbitrary location. Therefore, the LDOS accounts for the existence of the antenna and is an extent of its properties. In the absence of an antenna in free space, we achieve ( ) Purcell observed the dependency of the amount of atomic decay on the indigenous atmosphere in 1946 [33]. Since then, it has been used for different systems, such as near interfaces of molecules [34] or atoms in cavities [35,36]. The adaptation of atomic decay rates has a foundation in the interface of the atom-distinct secondary field. This distinct field attains the rear of the position of the atom after scattering in the indigenous surroundings. The transition frequencies and energy states are also infected by this back-action [37,38].

Antenna Impedance
According to circuit theory, the antenna resistance can be calculated as Re{Z} = P/I 2 . In an optical antenna, there is a governing dipole rather than a physical current which is more suitable for expressing Z according to the current density, ~ω j i p, as a replacement for the current, I. The antenna impedance, thus, can be defined as in [32] by the expression: Therefore, the antenna resistance Re{Z} can be linked with the LDOS. The unit of antenna impedance is Ohm per area in place of the typical Ohm. Here, Z is mutually dependent on the position ro and alignment np of the dipole. According to Greffet et al. [32], the stored energy can be found by the imaginary part of Z.

Antenna Efficiency
A basic problem in antennas is demonstrated in Figure 2. This figure contains dipoles p1 and p2, which are represented as a transmitter (Tx) and receiver (Rx). Here, the function of the antenna is to boost the Tx to Rx transmission efficiency, which can be achieved by raising the Tx radiation, for which a suitable figure of merit is the antenna efficiency and this antenna efficiency can be found as in [1]: where P is the total antenna dissipated power and Prad and Ploss means radiated power and power loss, respectively.

Directivity
The capacity of focusing the radiated power into a definite route is known as the directivity of the antenna, which represents the density of the angular power in relation to an isotropic radiator. The improvement of the efficiency of transmission can be accomplished by guiding the radiation towards Rx. Directivity is a measure of the proficiency for this system which can be represented as [1]: where both θ and ϕ denote the direction of observation and p(θ, ϕ) denotes the angular density of power.

Gain
Antenna gain is the result of the combination of antenna efficiency and directivity. The definition of antenna gain is similar to that of the directivity, but here the normalization is done in comparison with power P instead of the radiated power Prad. It can be mathematically represented as [1]: Directivity and gain are generally calculated in decibels. As isotropic perfect radiators are impractical, a more realistic approach is to state an antenna of known configuration. Then the comparative gain can be demarcated as the fraction of the gain in a specified direction to the gain of a reference antenna in a similar direction [28]. Bouhelier et al. recently described the relative gain of optical antennas, using the dipole-like radiation from single nanoparticles as a reference [39,40].

Reciprocity
Reciprocity makes it possible to trade off the sources and fields shown in Figure 2 to provide p1 · E2 = p2 · E1, where E1(E2) is the field of dipole p1(p2) calculated at the position of p2(p1). As a result, a noble transmitting antenna is also a noble receiving antenna. The reciprocity leads towards a correlation of emitter's stimulation rate (Γexc) with the impulsive discharge rate (Γrad) for a two-state quantum emitter which can be presented as [28,41]: where the meaning of the superscript "o" is the nonexistence of the antenna, while the subscript "θ" specifies the nature of polarization; explicitly, the points of the electric field in the direction of unit vector θ. An alike equation can be formed for polarization in the direction of ϕ. It is interesting that the excitation in high directivity direction allows Γexc to be boosted more intensely than Γrad. From Equation (6), it is clear that the relation of Γexc and Γrad due to the existence of the antenna is proportionate and has been included qualitatively in several studies [22,[41][42][43].

Antenna Aperture
Antenna aperture is another significant antenna parameter, which is similar to the absorption cross-section σ. Let, a dipole with σo cross-section as an Rx and the Rx is not connected to an antenna, np is a directional unit vector, and Eo is the receiver incident field. The receiver field increases to E after its connection with the antenna and antenna aperture (or absorption cross-section) can be found as Therefore, the antenna aperture changes with the indigenous intensity improvement factor. Many studies have revealed that 10 4 -10 6 intensity enhancements are possibly attainable [44][45][46] and therefore, a layer of molecules (all molecules are attached to an optical antenna) situated 0.1-1 μm apart are capable of absorbing all of the incident radiation for distinctive molecules with 1 nm 2 free-space cross-sections. Obviously, this evaluation has a limited validity because it overlooks the coupling among the antennas.

Effective Wavelength
In case of the radio wave antennas, the radio frequency (RF) counterpart of optical antennas, the wavelength of the incident radiation, λ relates to its design rules. For instance, the length of a half-wave antenna L is λ/2, and the separation between elements of a Yagi-Uda antenna corresponds to some fraction of λ [47,48]. Therefore, the scaling of RF antenna design from one wavelength to another is very straightforward because of its proportionality to λ. Conversely, this scaling is not applicable to the optical regime where the diffusion of emission into metals has to be considered. The delay after supplying the driving field to get the electronic response due to a finite electron density results in the skin depth and this skin depth is usually bigger than the antenna element diameter. Hence, the electrons of a metal react to an effective wavelength λeff instead of the wavelength λ. This effective wavelength can be determined as [49,50]: where n1 and n2 are geometric constants and λp is the wavelength of the plasma. Applying the new wavelength calculation rule, the wavelength of a half wave antenna for an optical regime becomes λeff/2 instead of λ/2. The difference between λ and λeff is influenced by the geometric constants, but is normally between 2 to 5.

Conductivity of Antenna Materials
As the conductivity of metals drops considerably when the diameters become less than 5 nm, metals are possibly not the best selection for antenna elements. In the case of diameters of less than 5 nm, carbon nanotubes are superior conductors than metals [51]. Therefore, on a small scale, carbon materials have wide possibilities to be the materials of choice for optical antenna elements [52].

Antenna Resonance
The resonantly excited nanostructures behave as optical antennas similar to RF antennas, especially in IR (infrared) that concentrate the energy of electromagnetic radiation to a confined volume of the sub-wavelength scale. Thus, nanorods, with µm-sized lengths L that show plasmonic resonances in the IR spectral range [53][54][55] are termed nanoantennas. However, the simple λ/2-dipole behavior known from RF antennas, where the relationship between L and the resonant wavelength λres is given by 2L = λres, does not hold for nanoantennas at optical frequencies. Moreover, the finite penetration depth of the light into the metal, and the non-negligible diameter D of the antenna lead to the modified relation [53]. 2 1 In this equation, λp denotes the plasma wavelength of the antenna's material, whereas the coefficients c1 and c2 rely on D and on the static dielectric constant εs of the surrounding medium. The basic assumption in this model is a high aspect ratio of the antenna (D << L) and the metal is described as a free-electron gas, as stated by the Drude model with negligible relaxation rate compared to photon frequencies. Since these conditions are adequately satisfied for gold nanorods with L/D > 10 in the mid-IR region, Equation (9) can delineate the resonance behavior of isolated or at least non-interacting nanoantennas.

Challenges and Outlook of Optical Antennas
There are a number of challenges to the broad use of optical antennas and they must be addressed before the technology becomes extensively usable.

Following the Analogy of RF Antenna
How far has the similarity between optical and RF antennas been achieved? Different nanofabrication tools have been extensively used for achieving such similarity. However, the RF antennas are driven locally at the feed gap, while the far-field was used to drive the initial optical antennas. For detection in the infrared regime, antenna-mediated transduction has been investigated, for instance, a slot antenna [56] or a dipole antenna ( Figure 3) [57]. Conversely, a huge number of enhanced RF antennas are needed to investigate in the optical regime. On behalf of directional recognition, customary loop and travelling-wave antennas are evident candidates. Specifically, the Yagi-Uda antenna scheme was practiced by a few researchers [58][59][60], who decided that the radiation at the feed component of an optical source can be totally focused in a single cone. Recently, experimental demonstration was made for the directional scattering from an optical Yagi-Uda antenna array [47,48]. Present RF antennas are greatly improved and enhanced in terms of dimension and bandwidth. The idea of using the fractal design of cellular phones in optics was first introduced by Stockmann [61]. Following this idea, modern fractal antennas can be designed for capturing and channeling a wide band of light to a single point, for instance the Sierpinski carpet [62]. In reality, the smallest dimension is a challenge, which sets a limitation on the operational frequency.
The directionality of the antenna can be controlled by impedance match regulation or by the use of a phased array, which permits quick routing of beams. An optical phase antenna has been suggested in [63] for attaining near-field focusing, but this type of phase controlling for a coupled optical antenna is still a challenge.

Impedance Matching
Impedance matching between source and antenna is one of the main challenges in designing an optical antenna. For a molecule-like source, approximately 1 MΩ reactive impedance can be acquired by considering the source as a capacitor, similar to a plate of dimension 0.2 nm × 0.2 nm with a similar separation [1]. Conversely, a typical metal nanostructure has an impedance commonly in Ohmic range and is exceptionally small, for instance ~3 Ω impedance is found for a half-wave linear antenna [1,28]. At present, the reimbursement process for this huge impedance gap is indistinct, but impedance matching is vital for effective connection between the source and radiation field. The initial platforms have been established [60,61,64,65], and for a half-wave dipole antenna, it has been presented that adjusting the feed-gap impedance allows proper tuning of the antenna resonance [66][67][68]. An atom's impedance cannot be fully imaginary due to the impulsive emission. The radiation losses must be explained by the real part of the impedance. For effective radiation, the excited state lifespan of an atom must be trivial. Therefore, the radiation resistance must be connected to LDOS [28,69]. Coupling to an optical antenna raises the LDOS, which makes the source (atom/molecule) a more proficient emitter [43,44,70,71].

Electro-Optical Conversion
In both of the RF and microwave frequencies, the main use of antennas is to transform electric currents into electromagnetic radiation, and vice versa. Conversely, the operational base of most optical antennas is "light-in, light-out". A small number of studies have reported the conversion of optical radiation into photocurrents by using antennae [57,72]. This transduction can be inspired by different high-frequency devices, like whisker diodes for infrared [15,17] and from photon radiation in scanning tunneling microscopy [73,74]. This type of infrared whisker diodes are able to deliver a pathway to Rx antennas [18]. The main difficulty in infrared whisker diodes is the metal-oxide-metal junction's capacitance-dependent high frequency cut off [56]. In contrast, scanning tunneling microscopy can decay radiatively according to the stimulation of surface plasmons by electrons [75,76]. However, evolving nanofabrication technology makes it promising to work with much reduced lengths, and it is anticipated that this type of conversion (electro-optical) by using the coupling of electron and plasmon will be achieved at some point.

Selection Rules
Here, we have already indicated little significant dissimilarity between optical antennas and RF antennas. For instance, the penetration of radiation into metals at optical frequencies cannot be ignored, and as a result, optical antennas react to the effective wavelength instead of wavelength of the inward radiation. Additionally, metals do not behave linearly in the optical regime, which creates the possibility of mixing and converting dissimilar frequencies. Lastly, but most significantly, the local fields near the optical antenna have spatial magnitudes that come near the length scale of the molecular quantum wave functions. This forms communication channels that are illicit by the selection rules of typical electric dipoles [28]. Moreover, the solid localized field near an optical antenna increases the momenta of photons up to the order of electron momenta in matter, and therefore increases the conventionally momentum-illicit alterations. Though interactions in the near-field beyond the rules of dipole selection are subject to theoretical studies [77], experimental verification is still needed.

Reproducibility and Repeatability
The progress of an optical antenna application depends on the capability of fabricating it with adequate material properties and precision on a nano-scale. Usually, the bottom-up technique has delivered good-quality crystalline metallic nanoparticles of manageable form and dimensions of a few nanometers. This type of colloidal nanoparticles display high-grade resonances (Q > 10) overcoming the restrictions of the dielectric properties of metals [78]. Additionally, triangle-, star-, core-shell-, and pentagon-like particles have been manufactured [79,80]. Although the fabrication (Figure 4) of antenna probes has been accomplished by selecting distinct or manifold gold nanoparticles, the composition of a designed antenna geometry by arranging the colloidal particles is still a challenge [70,71,81,82]. All of the replication is fundamentally dependent upon the skill and serenity of the researcher. Nowadays, prolonged optical antenna arrays of Au or Ag nanoparticles are usually fabricated through electron-beam lithography techniques [83]. Though 10-20 nm accuracy was achieved (Figure 4), the metals usually remain polycrystalline with 10-30 nm grains, which usually results in low quality (Q ~ 5) resonance, and thus, just one or two grains can affect the properties of specific antennas.  (c) Fluorescence from a metallofullerene [81].
For more complex 3-D structures, focused ion beam milling is an suitable substitute for electron-beam lithography [33]. Moreover, reproducibility is a matter of patience. Though antenna nanofabrication methods have advanced swiftly, the foremost difficulty at the present is how to provide for coupling between the antennas and the active optical sources, like nitrogen-vacancy centers, molecules, and quantum dots. For identifying the hot spots, post-selection can be used and thus the enhancement of antenna magnitude became pragmatic [44], where actual molecular antenna structures remained unidentified. Conversely, pre-patterned models are used in colloidal antennas [72]. In this case, more straightforward control has been attained by scanning antenna probes. Here, both of the comparative location and alignment are fully controlled, which gives an autonomous perception on the excitation improvement [41,70,71]. The perfect direct nanopositioning of particular quantum emitters in conjunction with complete control of the position and alignment corresponding to the antenna are the real test for future practical applications. Based on these physical properties, researchers have made many applications of optical antennas for different fields, such as nanoscale imaging and spectroscopy, photovoltaics, light emission, coherent control, etc. Here, in this review, we only discuss optical antennas in the field of biosensors.

Optical Antennas as Biosensors
For twenty years, broad research has been done in the field of biosensors for their substantial uses, such as in detection of DNA, different types of cancer and biomolecules. Therefore, an evolving technology in present biosensor research is optical antennas for label-free and real-time molecular recognition. Nowadays, numerous researchers are studying making bio-sensing optical antennas. As optical antennas are governed by the rule of plasmonics, these studies are divided into two different types, according to the surface plasmons involved: (i) surface plasmon resonance (SPR) biosensors [84] and (ii) localized surface plasmon resonance (LSPR) biosensors [85].
Surface plasmon polaritons (SPP) are the basis of SPR biosensors. SPR is a charge-density oscillation of SPP that might present at the interface of two media with dielectric constants of opposite signs, for example, a metal and a dielectric. The charge density wave is associated with an electromagnetic wave, the field vectors of which attain their maxima at the interface and decay evanescently into both media. The simplest and low cost SPR configuration is the Kretschmann configuration ( Figure 5a). Although SPR biosensors offer very high sensitivity, they are not adequate for small biomolecules in low concentrations. Recently, a new modulation technique has been employed, which surpasses the sensitivity of the standard SPR sensor. Therefore, these types of biosensors must be improved for practical use. LSPR biosensors are based on localized surface plasmon (LSP) phenomena [85,86]. At the plasmon resonant frequency, the optical extinction of a nanoparticle become maximum, which is dependent on the adjacent medium's refractive index and the nanoparticle's size and shape. Variations in the refractive index of the adjacent medium are noticed by the changes in peak wavelength of the LSPR. This peak wavelength change is found by measuring the spectral extinction, as displayed in Figure 5b. The bigger nanoparticles are more sensitive [87], but their peaks are stretched due to radiative damping and multipolar excitations.

SPR Biosensors and Present Research
SPR biosensors are able to detect biomolecular interactions directly without any labeling and thus allow real-time assessment of the kinetics and concentrations of analytes, and their thermodynamic binding parameters. Different features of familiar SPR sensors have been listed in Table 1.  Now, we are going to present different applications of SPR biosensors and their current research status.

Insecticide Detection
The U.S. Environmental Protection Agency (EPA) establishes the higher limits of acceptable concentrations for most pesticides; for example, the limit of concentration for atrazine is 3 ng/mL and for simazine it is 4 ng/mL [110]. On the other hand, the European Union permits up to 0.1 ng/mL and 0.5 ng/mL, for those two insecticides, respectively. A list of uses of SPR biosensors for pesticide detection is provided in Table 2.

Explosive Detection
Nowadays, there is a vast demand for optical sensors for detecting different explosives. In Table 3, a list of SPR biosensors for explosive detection is provided.

Environmental Pollutant Detection
Today's swift industrial development demands the pollutant detection for all livable atmospheres. Some SPR biosensor-based pollutant detectors along with their range is given in Table 4. Cadmium, zinc, and nickel 0.1 ng/mL in buffer Biacore SPR [131] Copper ions (Cu ++ ) 0.063 pg/mL-6.3 µg/mL [132] 3.

Bacteria Detection
Normally, bacteria detection using SPR biosensors is challenging for several causes [133][134][135]. One of these challenges is the smaller penetration depth of a 100 nm SPR biosensor in comparison with the usual bacterium size of 1-5 µm. As a result, most of the cell binding takes place in the exterior of the transitory field. An additional cause is a smaller refractive index difference of the detection environments, such as the bacterial cytoplasm and aqueous environments. Finally, the inadequate amount of diffusion of bacteria in the direction of the biorecognition elements on the surface of the sensor. In spite of these challenges, a number of researchers have been working in this field. Table 5 shows information on SPR biosensors for detecting harmful bacteria. Staphylococcus aureus 10 5 cfu/mL Reichert SR7000 SPR [147] Listeria monocytogenes Subtractive inhibition immunoassay 10 5 cfu/mL in less than 30 min Biacore 3000 [153] Listeria monocytogenes 10 5 cfu/mL Biacore 3000 [154] 3.1.

Virus Detection
Numerous researches have been conducted in virus detection using SPR biosensors. Table 6 provides some examples with details.  Human immunodeficiency virus type 1 (HIV 1) Using specific hybridization of immobilized biotinylated HIV-1 Biacore 1000 [158] 3.1.6. Toxin Detection Though most toxins can be detected instantly at higher density, normally for lower concentration, the sandwich assay is used. The recognition of bacterial toxins using SPR biosensors mainly relies on toxins' molecular weight. Table 7 provides information about some SPR biosensors for toxin detection.

Allergen Detection
Nowadays, the increased consciousness of allergens has led to rising calls for consistent, fast, and sensitive locating and measuring devices for allergens. Use of SPR biosensors for allergen detection is shown in Table 8. Histamine (3-imidazole) 3 ppb SPR 20 [174] Pollen of perennial rye grass Biacore [175] 3.

Biomedical Analyte Detection
Many SPR biosensors have been used in the field of biomedical analyte detection. Table 9 presents a list of these. Insulin growth factor-1 Milk Direct immunoassay 0.5-1 ng/mL Biacore [186] Human chronic gonadotropin (hCG) Buffer 0.05-1 µg/mL [187]

LSPR Biosensors and Associated Surface Enhanced Phenomena
A number of research groups are now trying to find alternate approaches for optical bio-sensing using the surprising optical characteristics of different nanoparticles. Nanoscale biosensors can be achieved with the shifts in LSPR [195][196][197]. Generally, LSPR biosensors work in a similar way as SPR sensors by transferring small refractive index variations into an assessable wavelength shift as follows [198,199]: Here, m is the sensor's refractive-index sensitivity, blank n and sample n are the refractive indexes of the bulk and sample environment, correspondingly, sample d is sample layer's effective thickness, and ld is the decay length of the characteristic electromagnetic field related to the sensor. Some differences between SPR and LSPR sensors are shown in Table 10. Other optical phenomena related to LSPR are also enhanced, for example surface-enhanced Raman scattering (SERS) and surface-enhanced fluorescence (SEF), as a result of the indigenous electromagnetic field enhancement nearby the nanoparticle [200]. Some applications of LSPR biosensors and associated surface phenomena are given below.

Wavelength-Shift Based Application
The responsiveness of LSPR sensors in a dielectric atmosphere is very high, which is beneficial for the exposure of conformational variations and molecular binding events, and are able to provide both kinetic and steady-state information. The wavelength-shift of LSPR has been employed as a conversion scheme for investigating the binding interactions of molecules. Table 11 shows the sensitivities of different LSPR biosensors in the nanostructure range.

Nano Plasmonic Molecular Rulers (PMRs) and PRET Biosensors
PMRs permit label-free estimation of different DNA and protein dimension and gap variations, instantaneous dynamic quantification of nucleic acid-protein ligand interactions, and confirmation of the existence of enzyme motion. PMR has a substantial advantage for durable kinetic studies in contrast to the Förster resonance energy transfer (FRET) method, as the conductive nanoparticles do not blink or photobleach [213]. Moreover, in contrast to FRET, which examines the binding events inside a 1-10 nm range, PMR deals with a separation space of up to 70 nm.
Ag or Au nanoparticle-based dimers have been employed for the measurement of DNA span together with hybridization kinetics. The space between two nanoparticles can be found based on the plasmonic coupling [214]. An additional study was dedicated to the study of nuclease activity, and wavelength shift was detected as a result of the variations in the dielectric constant with variations in DNA dimensions. In general, a 1.24 nm swing per base pair was detected [215]. Overlapping of the nanoparticle spectra and the absorption spectra of molecules supports the PRET, which results in spectral quenching as exposed in [216,217]. Additionally, a delicate and critical PRET basis sensor has been developed for Cu 2+ ions [218].

Nucleic Acid Hybridization Assays
Different bio-sensing and bioassay methods have been introduced by using nanoparticles for the recognition of the interactions of proteins, DNA hybridization, and different molecular actions. Huh et al. [219] identified DNA hybridization by using 50 nm Au nanoparticles in a microfluidic device that is operated by single stranded DNA (ssDNA). Here, passivation of nanoparticles was done using 6-mercapto-1-hexanol to reduce generic adsorption. Intended modifications of nucleic acids by hybridization and tetramethylrhodamine (TAMRA) brought the color into the near vicinity of nanoparticles (Figure 6a). In recent times, a sandwich assay using SERS for the purpose of DNA hybridization has been reported [220]. A capture ssDNA layer comprising thiols was restrained on a Ag nanoislands surface (Figure 6b,c). As a result of the use of nanoparticles for execution of the sandwich assay, the target recognition becomes enhanced from 1 nm to 0.4 fm.
Studies on molecular beacons were accomplished by modifying the DNA hairpin configuration through a Raman lively molecule [221][222][223]. This hybridization procedure interrupts the outline of the loop, which reduces SERS signal by growing the space of the Raman lively particle (Figure 6d).

Protein Recognition Assays
The recognition and investigation of proteins is usually accomplished using nanoparticle combinations, Raman labels, and surfaces of the nanostructure [224]. Pavel et al. [225] have studied small proteins, which after contraction have two cysteine moieties at axially reverse locations, performing as ligands. With the existence of Ag nanoparticles, FynSH3 proteins prompted the gathering of aggregates and dimers, which consecutively delivered a boost for SERS. The connection points are called "hot spots" due to the inherently small (2.3 nm) protein size. This research demonstrated the implication of protein location regarding the region of "hot spot". Formation of aggregates was also studied by Maher et al. [226] for different disease-specific enzyme recognition. Au nanoparticles were functionalized with N-fluorenyl-9-methoxycarbonyl (FMOC)-terminated peptide. The π-π connections among FMOC groups are the driving force of the aggregation and these could also be used as reporters with SERS. The prospective limit of detection for this scheme was likely as low as 10 −13 M, while the trial data attained a LOD of 10 −11 M, which is less than the essential range for biomedical uses. Au nanoparticles functionalized with thrombin were identified on heparin-adapted glass slides and anti-thrombin III [227], where a 10 −13 M LOD was found at a SN −1 ratio of 3.

Raman Labels
Different nanoparticles boost molecular Raman signatures while they themselves are in close vicinity. To increase the sensitivity, the coating of the huge amount of reporters has to use every nanoparticle. Gellner et al. [228] reported that a whole self-assembled monolayer (SAM) produces nearly 22 times stronger SERS in contrast to sub-monolayer coverage. SERS label confinement using organic polymers [229], proteins [230], and silica shella [231], increases solidity, removes the desorption probability of Raman reporters and reduces the generic adsorption. Encapsulated SERS assemblies, despite bigger dimensions, deliver a fabulous improvement in the intensities of SERS [232]. The conjugation of SERS labels with antibodies delivers recognition specificity and selectivity and is used in the immunoassays based on SERS [233,234]. A graphical demonstration of an immuno-histochemical assay using a SERS label is presented in Figure 7. Multiplexing was accomplished by these types of Raman labels with a dye range, and this technique is utilized for DNA recognition [232,233], and study of the interactions between different proteins and molecules.

Intracellular Detection
Nanoplasmonic elements have been exploited for imaging cells, probing the interaction of drugs and cells, and photo-thermal treatment. Various nanostructures comprising Au nanospheres, nanoshells, and nanorods have been employed as thermal converters for initiating permanent impairment of different cancer cell lines [235].
Though biosensing inside cells with different LSPR biosensors is at the initial stages, these techniques are effective for diagnostics, drug distribution and checking the efficiency of therapies. Medication effects on the plasma membrane of active cells for antitumor drugs can be observed with a sensitivity of 10 −10 M by SERS [236]. Kniepp et al. [237] studied the diffusion of drugs through the cell membrane with the use of Ag and Au nanoparticles that were supplied by endocytosis. Anticancer drug diffusion through cancerous cells was observed by Ag-layered silica beads in [238]. A SERS probes delivery method was revealed for the cell nucleus with a Raman reporter [239]. Breast cancer detection from blood samples has been done with a mixture of nanoparticles overcoated with epithelial and anti-her2 antibodies coupled with Au nanoparticles which possibly deliver SERS [240]. This technique delivered good sensitivity and high specificity using whole blood samples.

Challenges of Designing Optical Bio-Sensors
Although the future of optical biosensor technology depends on innovative researchers, the transition rate to the user community will be measured by various non-technical issues [241]. Moral apprehensions were also stated with respect to the genetic data usage and nanomaterial protection and such types of concerns about humans will eventually drive regulation. Different social considerations over problems, for instance depletion of resources will also force prime concerns for system strategy in addition to application ranges. The researchers who are working on the development of optical biosensors have a prime opportunity to integrate new knowledge into existing systems. The only restrictions appear to be the capability of integration of the elementary and innovative information with other disciplines, to gather skillful associates to support the effort, and to obtain economic and physical assets to investigate biosensors in the optical regime. Moreover, we need to study the critical users, consistency of the produced data, and the reaction to that data. We have listed below some points which we have to give emphasize before starting the design of optical biosensors: France has identified some infrastructure constraints (Table 12) [241,245].  The use of periodically corrugated metal-dielectric interfaces is a technique to overcome the wave-vector mismatch. The diffracted orders from the periodic corrugation have wave vectors larger in magnitude than those of the incident light. A light beam is directed towards a medium in which the surface has a spatial periodicity comparable to the wavelength of the incident light. The incident beam is diffracted, and the components of the diffracted light whose wave vectors coincide with the SP's wave vector get coupled to the SP. Efficient coupling is provided to both air-metal and substrate-metal SP modes of a metal film, if the film thickness and the grating corrugation depth are properly allied.
The main benefit of grating-coupled biosensors is that they can be produced by mass replication technologies, such as injection molding and hot embossing. These technologies have have facilitated to produce low-cost and high-throughput biosensing platforms for label-free monitoring of biomolecular interactions [110,246].

Human Exposure to Nanotoxicology
Conventionally, toxicology addresses hostile effects of poisoning due to chemicals on human beings, animals and the environment. Figure 9 represents the graphical view of the advantages and disadvantages of nanoparticles. Figure 9. Nanoparticle administration paths and their advantages and disadvantages [247].
Many researchers have concentrated on developing nanoscale materials which might hold probable reasons for toxicity [247][248][249][250][251] and researchers must pay attention to this fact during the design and implementation: (i) All of the nanostructures have their own optical, electronic, and magnetic properties associated with their dimensions. The breaking of these types of structures may cause a poisonous outcome. (ii) The surfaces of the nanostructure take part in many oxidative and catalytic reactions. If these responses produce cytotoxicity, the deadliness becomes superior for greater surface-tovolume ratio. This surface-to-volume ratio increases when the diameter of a spherical particle becomes smaller and due to this increased surface area, the chemical reactivity is increased. This is most significant for nanobiological relations. (iii) Many of the nanostructures have composites with well-known harmfulness and in consequence, the disintegration of these composites will possibly cause alike toxic reactions to the metals or composites themselves. Table 13 highlights the applications and concern surveys for the most frequently used nanostructures. (iv) Traditionally, toxicology is related to the conception of dose and reaction to a dose. In nanotoxicology, the evaluation of appropriate and realistic dose ranges is essential for achieving significant results from public health risk assumption experiments. Therefore, the nano-toxicologist ought to test the toxicity of nanoparticles on the basis of practical doses in spite of the impractically high doses for achieving a biological reaction. (v) The size of nanoparticles is at the same scale of protein molecules and they are capable of interfering with the signaling systems of cells. They also can interact with proteins, by chaperone-like action or by altering their configuration. This type of protein missformation leads to neurodegenerative infections. Inspecting probable missformation of proteins and macromolecules is vital for nanotoxicological research.  [252,257] To avoid the abuse of nanoparticles in humans, it is necessary to carefully monitor the above described toxicity reasons through the whole process of biosensor design to implementation. Along with this, we have to use the toxicity and safety information of nanomaterials for risk evaluation and management. Outcomes from quick testing techniques must be made accessible for the assessment of the toxicity of nanoparticles to ease the nanoparticle risk assessment process. The conclusions of a cellular test scheme and prudently certified in vitro trials and high-throughput approaches using proteomics and genomics have to be made accessible for risk evaluation [259]. Some legal approaches of nanoparticle risk assessment are given below: The environmental and health aspects of nanotechnology have only been concerns for the last decade or so since 2003, as there were no scientific peer-reviewed articles about such issues [260], and up to August 1, 2005, there were less than 10 papers on the safety evaluation of nanomaterials [261]. However, the article [254] attempted to trace the historical origin of toxicological research relating to nanoparticles and mentioned that the toxic effects of some nanoparticles were reported even two decades ago.
One of the serious tensions before the introduction of nano-enabled products is that many stakeholders compare this with that of genetically modified foods, which faced unprecedented challenges in the recent past. A significant portion of the consumers refused to welcome these foods. Therefore, risk assessment processes for nanomaterials are important to convince the consumers about their safety.
The risk and safety concerns of nanotechnology are almost contemporary with its emergence. The Organization for Economic Co-operation and Development (OECD) has identified seven types of risks associated with nanotechnology, i.e.: (a) business risks, due to marketing of products; (b) intellectual property protection risks; (c) political risks due to economic development of countries; (d) privacy risks due to unlimited use of sensors; (e) environmental risks due to nanoparticle release; (f) safety risks of workers and consumers; and (g) futuristic risks, e.g., human enhancement and self replicator [262].
The general practice is that both the hazard and exposure are to incorporate into a risk assessment paradigm, consisting of Hazard Identification, Hazard Characterization, Exposure Assessment and Risk Characterization in order to make appropriate risk management decisions [263] opined that the four steps of traditional chemical risk assessment procedure, i.e.: hazard identification, dose-response assessment, exposure assessment, and risk characterization may not be suitable to be used for assessment of nanomaterials. The Scientific Committee on Emerging and Newly Identified Health Risks of the European Commission (SCENIHR) suggested to improve the risk assessment process to address all hazards relating to the exposure of nanoparticles and for developing an improved system, there is no alternative to conducting more research. According to SCENIHR, such a system should target the routine determination of physical and chemical properties of nanoparticles, measure the representative exposure to free nanoparticles in the environment, i.e., in air, soil and water, assess whether nanoparticles exacerbate pre-existing medical conditions and the movement of nanoparticles inside living systems.
The issue of risk assessment is crucial in assessing the health and environmental impacts of nanoparticles and the article [264] shared that the following things should be considered in risk assessment of engineered nanoparticles: "(a) exposure assessment of manufactured nanoparticles; (b) toxicology of manufactured nanoparticles; (c) ability to extrapolate manufactured nanoparticle toxicity using existing particle and fiber toxicological databases; (d) environmental and biological fate, transport, persistence, and transformation of manufactured nanoparticles; and (e) recyclability and overall sustainability of manufactured nanomaterials" [264].
In order to assess the possible risk out of nanoparticles, it is imperative to analyse the whole life cycle of the nanoparticles, which includes the understanding of the processes and materials used in their manufacture, the possible interactions between the product and the individual or the environment during the manufacturing stage or during the disposal stage. Simultaneously, this is also important to consider the defense system of the body of the persons who will be dealing with the nanoparticles (U.K. Nanotechnology Working Group, The Royal Society and the Royal Academy of Engineering. London, July 2004, at pp. [35][36]. In [259], the conclusion is that though there is no need to alter the current overall approach to the risk assessment of chemicals, the data gaps of engineered nanomaterials (ENM) risk assessment include: "(1) ENM aerosol standards and agreement on ENM key metrics; (2) dependable exposure scenarios, affordable monitoring technologies, exposure data and models; and (3) biomedical data on ENM translocation and toxicity, and associated testing strategies (which must be linked to the exposure scenarios)" [259].
The definition of nanomaterial seems to be crucial in the risk assessment process. The Joint Research Center of the European Commission assessed the working definitions available in different countries like the UK, USA, Canada, Australia, Denmark and in different organizations, like International Organization of Standardization, European Union, European Committee for Standardization and found that there are different scales used to define nanoscale and the nanoscale ranges from "1 nm to 100 nm", "up to 200 nm (in two or more dimensions)", "between 1 nm to 100 nm", and "less than 100 nm" [265]. However, there are also counterarguments where it is claimed that for nanotechnology, the definition of "one size fit for all" may not be suitable and therefore, nanomaterials should be considered case by case [266].
European Food Safety Authorities are the first international body that included the determination of exposure and toxicity testing strategy. The Scientific Committee on Consumer Safety of the European Commission has released guidance for the assessment of nanomaterials in 2012.
In [267], the author developed a tool named "NanoRiskCat" to assist companies and regulators to assess hazards and exposure potential of consumer products containing engineered nanomaterials. This tool is quite helpful to assist stakeholders in making decisions about more information as to exposure and effect, which will further assist to decide on the safe use of nanomaterials.
An interesting development is that based on control banding, a system to assess risks in pharmaceutical industries, an online risk-banding tool, i.e., Stoffenmanager Nano (version 1.0) was developed for the assistance of the employers and employees to prioritize health risks arising out of exposure to manufactured nano-objects (MNOs) [268].
In [269], the authors advocated for an integrated approach to specific risk analysis at work. They revealed some gaps in the whole process, i.e., "restricted information, problems in relating nanotechnologies and production of nanomaterials to specific areas of application, efforts required to assess the hazards posed by nanomaterials in realistic exposure conditions, ethical issues about nanotechnology in the workplace expected to arise from today's knowledge about the hazards of nanomaterials and the risks they may pose to workers" [269]. Therefore, "an integrated approach to research, cooperation, and communication strategies is indispensable, if we are to direct our efforts towards responsible and sustainable growth of nanotechnologies" [269].
Instead of setting the agenda to assess the risks of nanomaterials in general, a case by case approach is suggested in a number of researches. As a result, different countries have conduced risk assessments of different nanomaterials, especially those which are mostly used in consumer products and concluded that within the existing knowledge and research findings, such nanomaterials are not injurious to human health. For example, the National Institute of Advanced Industrial Science and Technology of Japan completed the risk assessment of titanium dioxide (TiO2), fullerene (C60) and carbon nanotubes (CNT). Similarly, the Australian National Industrial Chemicals Notification and Assessment Scheme (NICNAS) confirmed that TiO2 is not normally toxic.

Prospective and Conclusions
Optical antennas are the prevailing tool for the manipulation of light on a nanometer scale and they are also capable of delivering optimum control over transduction in the far-field region. Present optical antenna research is being motivated in particular by developments in nanofabrication technology and RF antenna analogies. Though various antenna conformations are currently being appreciated in the optical regime, it is going to be fascinating to observe how different antenna parameters, such as the impedance matching, are going to be redefined for different types of optical sources, like atoms and molecules. Optical antennas unite the quantum methods and photon sources by including fascinating new physics, for instance the breach of selection procedures and unconventional ways for robust pairing. The ideas of focused radiation and focused reception can be pragmatic to the photon emitters. Once the techniques of nanofabrication have been become mastered, a variety of applications will appear, including controlled single-photon sources for quantum information, light harvesting, energy conversion, efficient biosensors, data storage, nanoscale optical circuitry and optical imaging beyond 10 nm resolution.
This review has emphasized the principle and applications of optical biosensors. An appreciation of optical antenna basics offers chances of tuning and controlling the optical performance. Optical antennas have a robust reliance on the shape, size and composition of the nanoparticles which deliver an enhancement of biosensors' sensitivity. A wide range of investigations on optical antenna are now dedicated to making substrates, which afford solid improvements of the EM field and deliver information about attaining control of optical properties by controlling the physical factors of nanoparticles. In spite of the different challenges for practical implementation of optical biosensors, the technical works show that importance in optical biosensor improvement continues to increase at a tremendous pace.