Pathogens are defined as foreign antigens that have adverse effects, such as (1) producing toxins, (2) penetrating tissues, (3) colonising tissues, (4) intercepting nutrients, and (5) immunosuppressing the host [1,2,3]. Particularly, contagious pathogens (e.g., those that cause bacterial urinary tract infection, malaria, influenza virus, dengue virus, human immunodeficiency virus (HIV), and severe acute respiratory syndrome coronavirus (SARS-CoV-2) are capable of human-to-human presymptomatic transmission [4,5,6,7,8,9]. Infection pathways with receptor-ligand interactions (RLIs) constitute a representative mechanism through which pathogens bind to their target [10,11,12,13]. Additionally, given that RLIs are biological events that occur in living cells, a high-sensitivity diagnostic sensor is required to detect pathogens [14]. Therefore, biosensing approaches must be constantly optimised to achieve better signal sensitivity.

To increase the sensitivity and selectivity between specific pathogens, recent studies have developed and evaluated novel diagnostic tools that incorporate pathogen binding mechanisms and antigen capture strategies, such as antigen–antibody and aptamer–ligand interactions [15,16]. Current diagnostic procedures are based on several clinicopathological analysis methods, such as gene sequencing, polymerase chain reaction (PCR), fluorescence methods, mass spectrometry, and enzyme-linked immunosorbent assays (ELISA) [17,18,19,20,21,22,23,24]. However, although these bioanalytical methods are highly accurate, each of them has unique limitations. For example, (1) the fluorescence method requires labelling the sample with a fluorescent probe, which adds complexity to the procedure; (2) the analysis of multiple samples using mass spectrometry is highly time-consuming; and (3) most of the aforementioned techniques are generally expensive and require sophisticated equipment and specialised processing, in addition to having a high response time [25,26,27,28,29,30]. To address these limitations, bioanalytical techniques based on optical biosensors have recently garnered increasing attention [31].

Particularly, plasmonic-based biosensors constitute a next-generation technology that facilitates the miniaturisation of biosensors, thereby enhancing detection throughput and reducing operating costs [32]. Pathogen detection can be based on a variety of plasmonic phenomena, including SPR, localized SPR, surface-enhanced Raman scattering, surface-enhanced infrared absorption spectroscopy, and surface-enhanced fluorescence [33]. Among these novel approaches, surface plasmon resonance (SPR)-based sensing techniques are a promising new means to detect pathogens quickly with high sensitivity, and therefore, this method has begun to replace other conventional diagnostic approaches. SPR is an electromagnetic surface wave that propagates parallel to the interface region. On the basis of this principle, changes in a given sample can be detected by monitoring the intensity of the SPR angle (θ) reflected after the plasmon wave of a medium with a low refractive index permeates [34,35]. SPR sensors have some minor disadvantages, such as the possibility of sample SPR peak deflection when a strong electromagnetic field acts on sensitive biological samples [36]. However, there are four main advantages that compensate for these drawbacks: (1) this approach is non-destructive and non-invasive—in other words, it does not cause any damage or physicochemical transformation to the sample; (2) it allows for the sensitive optical quantification of the interaction between molecules in real-time without the need for a separate probe using a radioactive or fluorescent material; (3) after the detection of the target analyte, the surface of the sensor can be reactivated and used repeatedly upon treatment with a regeneration solution; and (4) the lifetimes of nanoparticles can be flexibly controlled and extended through surfactant surface coating or chemical conjugation (e.g., hybrid nanomaterials) [37,38,39,40,41]. Additionally, this approach is rapid, accurate, cost-effective, requires no specialised training, and can be applied to detect a wide variety of targets [42,43,44]. Collectively, these advantages make SPR-based sensing techniques especially well-suited for real-time diagnosis both in clinical laboratories and in the field [45,46,47].

2. Graphene Oxide-Based SPR Sensor

The edge of the graphene oxide (GO) structure is rich in oxygen functional groups; it is a single-layer, two-dimensional carbon material [92,93]. GO has outstanding properties that show great potential for biosensing at the molecular level because of its abundant π-conjugation structure along with its large surface-to-volume ratio [94]. Therefore, GO has specific intrinsic features that enable the robust immobilization of biomolecules on the SPR biosensor platform.

A recent study reported the detection of Leptospirosis bacteria using a graphene-based SPR sensor [76]. The researchers coated a glass platform with graphene to effectively adsorb the bacteria. Rodent urea was then used as the main sample, and the analytical outcomes were classified as polyuria (high urine production) and oliguria (low urine production). Upon measuring the urea with the SPR platform, polyuria was defined between 128.8378 (deg/RIU) and 130.40 (deg/RIU), whereas oliguria was defined between 134.5118 (deg/RIU) and 131.0978 (deg/RIU) with different angles of reflection. The proposed graphene-based SPR sensor has been theoretically and numerically analysed to detect Leptospirosis-causing bacteria in rodent urine, suggesting that it can be utilised for the early detection of this disease with high sensitivity and accuracy. Although graphene can be used to increase SPR signals, additional efforts have been made to improve sensitivity and detection accuracy. To achieve this, researchers have begun to study barium titanate nanocomposites. Through the adsorption of barium titanate onto graphene, the proposed sensor achieved a higher detection sensitivity compared to a single graphene-based SPR biosensor. Furthermore, Z. Xia et al. proposed an SPR sensor based on barium titanate and also achieved high analyte sensitivity [95]. Another study detected Pseudomonas bacteria by mixing graphene and barium titanate [77]. The platform was fabricated by deposition of silver and graphene on a BK7 prism glass plate that refracts incident light, and the sensors were constructed in the presence or absence of barium titanate (BaTiO₃) nanocomposites. The authors reported that the maximum sensitivity, detection accuracy, and quality parameters of the sensor were 220 deg/RIU, 7.09, and 101.38 RIU⁻¹, respectively. Therefore, this SPR platform can sensitively detect Pseudomonas bacteria. Additionally, the authors demonstrated that the graphene thickness had a critical effect on the results even when the affinity of the layers changed, and the monolayer exhibited the best reflectivity curve.

Quantum dots are a nanomaterial with a high surface area and signal-to-noise ratio, and significantly improved data accuracy [96]. Additionally, combining graphene oxide with carboxyl functional groups and quantum dots greatly increases mechanical stability as well as the potential for SPR signal amplification [97]. A study from 2019 reported the detection of dengue virus E-protein by adsorbing cadmium sulphide quantum dots on an optical SPR platform based on graphene oxide [78]. The SPR platform with cadmium sulphide quantum dots was fabricated through the stirring method, and the matrix of the sensor was based on graphene oxide. The fabricated SPR platform achieved a detection limit of 0.001 nM for the dengue virus E-protein, which demonstrated its high sensitivity, and the estimable linear range of the sensor was 0.0001 to 0.01 nM. These sensing results suggest that graphene oxide can improve the performance of SPR biosensors and is thus a promising candidate in the field of biosensing.

The above-described observations demonstrate that graphene oxide-based materials have various advantages for pathogen detection. It has been proven that GO does not damage biological samples, nor does it interfere with the SPR effect. Moreover, because of the characteristics of GO, such as its easy handling and excellent analyte adsorption properties, highly sensitive GO-based SPR sensors are expected to be developed in the future.

3. Gold-Based SPR Sensor

Gold-based materials can be easily synthesised and are highly biocompatible. Additionally, these materials are structurally strong, and their optical properties can be fine-tuned depending on the conditions of the platform. Therefore, gold is widely used in biosensing chips [98,99]. Over the past decade, gold has been applied in various forms, and previous studies have demonstrated that this material exhibits strong red-shifting properties [100]. Furthermore, the strong radioactive properties of gold particles, such as absorption, scattering, and plasmonic field generation, can be widely applied to SPR biochips.

Bong et al. developed a gold-based SPR sensor for SARS-CoV-2 [79]. The sensor was fabricated by binding antibodies through chemical conjugation on the gold chip sensor, which was the base of this SPR platform. The limit of detection for an analyte isolated from serum was 1.02 pM (i.e., the sensor was highly sensitive). Additionally, the sensor only detected coronavirus and no other influenza viruses, thus demonstrating its high selectivity for the analyte due to the characteristics of the substrate. Moreover, the gold-based SPR sensor rendered results very quickly, making this approach uniquely well-suited for the accurate detection of diseases at an early stage.

Spherical gold nanoparticles have various advantages compared to gold films or random gold particles, such as their narrow size distribution, consistent shape and size, and easy handling. Particularly, the size specificity of gold nanoparticles results in a narrow standard deviation from the analyte signal during SPR measurement. Furthermore, antibodies that specifically recognise various chemical conjugates or analytes can be attached to the spherical particles, thereby taking advantage of the large surface area of the nanoparticles and enhancing the SPR signal. A study from 2017 reported the development of a surface plasmon sensor using an antibody on a gold-based platform to sense pathogenic bacteria [80]. The fabricated biosensor was based on the principle of strengthening the surface plasmon effect by adding gold nanoparticles (AuNPs) after binding the analyte on the gold sensor chip. The limit of detection for the analytes of this fabricated SPR platform was 8 × 10⁶ CFU/mL, and the sensor was specifically sensitive to the pathogenic bacterium Campylobacter jejuni. Furthermore, the AuNPs in the sensor increased the efficiency of analyte analysis, whereas the gold sensor chip enhanced SPR sensitivity. Interestingly, although the use of AuNPs enhanced SPR performance, the SPR effect was maximised when using other types of gold nanostructures, such as nano-urchins or nanospikes.

Similar to sea urchins, gold nanospikes are not perfectly circular but have a sharp and uneven surface. This unique structure could enhance the red-shifting of the surface plasmon resonance absorption peak, in addition to increasing the electromagnetic field at the tip of the spike structure compared to that of a spherical particle. A recent study developed a gold spike nanoparticle-based SPR sensor that detects SARS-CoV-2 in human plasma [81]. The researchers deposited the gold spike nanoparticles through electrodeposition onto a glass slide rather than gold in its bulk state. The fabricated SPR sensor exhibited a low limit of detection of 0.08 ng/mL (~0.5 pM) for SARS-CoV-2 in addition to being highly specific, as it did not react to three competing substances and BSA protein. This shows that the fabricated SPR sensor could be used for the early diagnosis of COVID-19 and can accurately recognise virus analytes without detecting other substances remaining in the human plasma.

Studies have demonstrated that bimetallic combinations can provide greater sensitivity, signal-to-noise ratio (SNR), and increased operating range values, all of which cannot be achieved with a single metal layer. A recent study sought to offset the shortcomings of single metal layer-based sensors and improve the performance of the SPR sensor by using a novel bimetal layer structure of Au and aluminium (Al) [82]. The researchers detected the DENV2 NS1 protein with the developed SPR sensor by laying a gold and aluminium bilayer on the SU-8 spacer. The fabricated SPR sensor recorded a limit of detection of 0.1 μg/mL for DENV2-NS1, and the linear range of the sensor was between 0.1 and 10 μg/mL. Interestingly, the limits of detection in both PBS and blood plasma were also within a similar range (i.e., as low as 0.1 µg/mL), suggesting that this approach can detect viruses directly in blood samples or liquid biopsies. This study demonstrated that a specific virus can be detected with high sensitivity through immediate screening in clinical samples.

According to a recent study, on-chip integration of electronics and plasmonics would allow for samples to be guided to relevant locations and sensed thereafter [101,102,103,104]. Particularly, gold-based plasmonic trapping technology could substantially improve the efficiency of broadband photodetectors and SPR sensors [105,106]. Gold nanostructures have a virtually infinite potential depending on how they are applied. These structures can be constructed using various methods, such as growing gold on a substrate, attaching it by chemical conjugation, or depositing it. Similarly, the ability of the sensor to detect the analyte can be fine-tuned. The implementation of gold in the SPR sensor enhances its surface plasmon effect, the most basic characteristic of metal, and causes no damage to the platform body, both of which are considered crucial advantages of this material.

4. Hybrid Graphene/Gold-Based SPR Sensor

Gold-based biosensors are currently preferred over other sensors based on other metals. However, according to a recent study, there is a risk of peeling if the gold layer coating on the platform surface becomes too thin [107]. Furthermore, another study reported that gold itself has little influence on the adsorption of biomolecules, which can lower the detection limit [108]. Additionally, sensors based only on graphene exhibit an excellent adsorption force for the detection of analytes, but their surface plasmonic effect is significantly lower than that of metal-based sensors [109]. To compensate for this limitation, several recent studies have developed biosensors based on a hybrid graphene/gold structure.

Omar, N.A.S. et al. developed a hybrid graphene/gold sensor to detect the dengue virus by thinly coating reduced graphene on a gold film [83]. The generated SPR platform was coated with gold in the form of a film through deposition on a glass substrate, and graphene was thinly coated through spin-coating. This hybrid sensor achieved a detection limit of 28 fM and a linear range of 0 pM to 10 pM, as determined with a dengue virus solution. A sample of 100 pM human serum albumin (HSA) was used for the selectivity test, and the fabricated platform reacted only with the dengue virus and showed low reactivity to HSA. These results demonstrated that the hybrid graphene/gold SPR sensor could potentially be used in health and environmental monitoring. In addition to gold films, gold nanoparticles, an important type of plasmon nanoparticles, possess excellent plasmonic properties and have thus begun to be implemented in a variety of applications. The use of particles maximises the surface area of the gold film and graphene oxide complex. At the same time, the spherical gold nanoparticles adsorb to the graphene to enhance adsorption and plasmonic performance [84]. An SPR sensor was developed by combining graphene oxide and spherical gold nanoparticles through sonication and stirring. The researchers detected A-type and O-type foot-and-mouth disease viral particles through the resulting substrate and achieved limits of detection of 100 fg/mL and 100 pg/mL, respectively. Because of the characteristics of the GO on the substrate, single-stranded DNA (ssDNA) can readily bind to the sensor through electrostatic interactions, and thus the limit of detection is 1000 times lower than that of conventional PCR. Therefore, combining pristine gold nanoparticles and GO can greatly improve the limit of detection of biosensors.

Many SPR sensors have been developed to detect analytes by adsorbing gold nanoparticles on graphene; however, this method is notorious for its limitations. Specifically, a recent study reported that the performance of the sensor was low because the adsorbed gold nanoparticles tend to agglomerate on the platform surface, and the absorbance curve is widened, thereby decreasing sensor stability [110]. To solve this problem, recent studies have explored the encapsulation of gold nanoparticles using graphene. For example, a related study used GO to encapsulate gold nanorods, thereby overcoming the aforementioned limitations [85]. Concretely, the researchers developed a GO-encapsulated gold nanorod-based SPR sensor to detect hepatitis B, and this platform was fabricated through sulphonylation. To increase the specificity of the sensor towards hepatitis B, HBsAg antibodies were decorated on the GO surface. The limit of detection for hepatitis B was 0.05 pg/mL, and the sensor had a wide linear range of 1–1000 pg/mL. The researchers then diagnosed HBV-infected patients with the fabricated platform and achieved a relative standard deviation (RSD) of less than 5%, thus demonstrating the potential applicability of the sensor in clinical diagnosis.