Guided Mode Induced Surface Phase Mutation for Enhanced SPR Biosensor with Dual‐Parameters Interrogation

Ongoing research on the sensitivity and integration of refractive index‐based biosensors has resulted in significant advancements. Here, the study presents an enhanced surface plasmon resonance biosensor that integrates imaging technology and features dual‐parameter interrogation (intensity and phase) with guided mode coupling. By depositing a silica‐waveguide‐layer on a metal‐layer, two‐mode coupling is established to generate a high Q resonance and induce a phase mutation. The sensing performance experiment demonstrated a phase sensing sensitivity of 1.1 × 105 degree RIU−1, Q‐value of the resonant peak up to 314, and figure of merit of 300 RIU−1, superior to most standard plasmonic sensors. An in‐line phase‐polarization modulation scheme combined with imaging technology is proposed to extract the resonant phase carrying refractive index information. Additionally, a pair‐prism module is designed to optimize the sensing system configuration. Meanwhile, dual‐parameters interrogation including the intensity and phase are demonstrated, which offers potential for complementary and multi‐sensing fusion applications. The intensity interrogation also shows a considerable sensitivity of 7.2 × 104 a.u. RIU−1. Furthermore, it is combined with microfluidic chip to detect of alpha‐synuclein protein closely related to Parkinson's disease, and the limit of detection can reach 300 pg mL−1 level, which indicated a considerable potential for high‐throughput diagnosis application.


Introduction
[3][4][5][6][7] SPR sensors offer label-free detection, low sample consumption, and high sensitivity.The sensing mechanism of SPR sensor relies on the coupling mismatch of the propagating wave vector caused by changes in the dielectric environment due to the binding of specific target molecules.This bonding empowers it with the capability of identifying specific targets, which provides opportunities for applications in biosensing and chemistry analysis.The coupling mismatch usually leads to changes in the optical properties of exciting light, such as amplitude, resonant wavelength, resonant angle, or resonant phase.10][11][12] Yet direct inspection of resonant phase is challenging.[15][16] Nevertheless, these proposals have some practical limitations due to the signal disturbance caused by the inherent non-common light path in the system, the difficulty in integrating with high-throughput chips, or the cost of expensive detecting instruments.
[19][20][21] While gold offers decent biocompatibility and chemical stability, it is relatively costly, whereas silver is cheaper but easily oxidized when exposed to air.24][25][26] As a result, the phase change at the resonance peak tends to be flat, yielding low sensitivity.In contrast, dielectric layers can support a guided mode when the wave vector matches. [27,28]Such modes, with low inherent losses, may have narrow resonant peaks with high Q-values.[31] Therefore, it may be a good solution that combines both metal and dielectric for this contradiction.[34] But those methods usually need numerous fabrication steps, tedious manufacturing processes, or bulky equipment and have difficulty in large-scale preparation.Besides, high Q-value sensing structure is also beneficial to the intensity interrogation sensor, owing to its intense energy coupling.
In this paper, we proposed a simple yet effective method to form the coupling between a guided mode and SPR mode by depositing a silica dielectric layer on the metal.By forming the two modes coupling on the interface, we obtained a resonant peak with an extremely narrow FWHM and high FOM.The FWHM of the resonant angular spectrum can reach 0.2°level, the Q-value is ≈314 and the FOM is ≈300 RIU −1 .Furthermore, a simple phasepolarization modulation method is proposed to extract the phase information without non-common light path and the beam split from external modulation.Two modes coupling induced a phase mutation, and the phase sensitivity is demonstrated to reach to 1.1 × 10 5 degree RIU −1 .Additionally, we demonstrated the dualparameters interrogation including the phase and intensity at the same time.Finally, we applied it in alpha-synuclein protein detection that was demonstrated to be closely related to Parkinson's disease and achieve a limit of detection <300 pg mL −1 .This system could also be integrated with an imaging system and microfluidic chip for high-throughput sensing.Our sensor design and scheme may play an important role in the field of biosensors, chemical analysis, or disease diagnosis.

The Mechanism and Simulation for Modes Coupling
The occurrence of the conventional propagation-type metallic SPR effect depends on the matching of the wave vector.Whilst the wave vector component of the incident light along the propagation direction matches the surface plasmon wave supported by the metal-dielectric interface, the plasmon resonance occurs, and the incident light couples to the SPR mode.However, the wave vector in free space is always smaller than the propagation constant of SPR mode, as shown below: Therefore, owing to unsatisfying the matching condition, the light in free space cannot directly excite the SPR effect at the metal surface.Some methods such as utilizing a total reflection prism, an optical waveguide, or a grating are needed to raise the wave vector of incident light to satisfy the matching condition.The light with the risen wave vector will satisfy the matching equation at a certain frequency (fixed incidence angle), or a certain incidence angle (fixed frequency).
The waveguide layer can support a guided mode with a wave vector larger than free space.The wave vector of the SPW (Surface Plasmon Wave) mode is also greater than the free space wave vector.Normally, the guided mode is affected by the dielectric constant and thickness of the waveguide layer.The SPW mode, as shown in Equation (1), is affected by the dielectric constants on both sides of the interface.By introducing a dielectric layer supporting the guided mode, we may make the two modes satisfy the matching condition and couple with each other.By adjusting the incident angle, the dielectric constant, or the thickness of the waveguide layer, the incident light can excite the two modes of coupling.The guided mode provides a specific mode selection, like a "frequency sieve".Thus, this "frequency sieve" enables a high Q resonant peak with a small FWHM, simultaneously accompanied by a sharp phase mutation.
To furtherly reveal the guide-SPR mode coupling mechanism, we first consider a simple structure like the construction in Figure 1a,d.This structure consists of a sensing layer at the top, a waveguide dielectric as the gap layer, and a metal layer supporting the plasmonic resonance at the bottom and the substrate is K9 glass.The refractive index of the top, gap, bottom, and substrate layers is n t , n g , n b , and n s , respectively.The thickness of the top, gap, bottom, and substrate layer is h t , h g , h b , and h s , respectively.Figure 1a also presents the microfluidic part combined with the sensing chip.The microfluidic serves as a carrier for the liquid sample.
The transfer matrix method (TMM) is used to calculate the propagation characteristics of the light in multilayer films.First, calculate the transfer matrix of each layer.The total transfer matrix containing the propagation information is calculated by the method of matrix multiplication in the wake.Then calculate the propagation parameter based on the total transfer matrix.
Based on TMM, we can obtain the transfer matrix of the ith layer: Here, means the wave vector in vacuum.h (i) means the thickness of ith layer.n (1) and n (i) mean the complex refractive index of first and ith layer, respectively. means the incident angle of beam.
Therefore, the total transfer matrix can be calculated by matrix multiplication: The complex reflection coefficient can be obtained following: r = |r| means the total reflection coefficient of multilayer films, and  = arg(r) means the phase value of reflected light.Thus, by selecting the materials, tuning the thickness, and adjusting the incident angle, we can control and trim the coupling between a guided mode and SPR mode.Meanwhile, the total reflection coefficient and the reflection phase value of multilayer film can be tuned.
As shown in Figure 2a, as a certain thickness of silica waveguide layer is introduced on the silver surface, the waveguide mode will couple to the SPR mode, and a narrow resonant peak will be formed at a certain incident angle (satisfying the resonant condition).The simulated angular spectrum shows that the resonant peak of two-modes coupling is narrower and has a smaller FWHM, compared with bare silver-and gold-based SPR (Figure 2a).Meanwhile, the narrow resonant peak is accompanied by a sharp phase mutation at the resonant point.In pace with the change of the refractive index in the sensing layer, the resonant phase will also change accordingly (Figure 2b).The silica waveguide layer is not only playing the role of introducing the coupling but also an air insulator layer.The existence of a silica layer blocks the direct contact between silver and air, which is beneficial to sensing applications.When we changed the thickness of the silica layer, it was found that not only the resonant angle but also the FWHM also increased accordingly (Figure 2c).As the increase of silica thickness, the intrinsic wave vector supported by it rises following.Thus, to meet the matching conditions, the incident light needs a larger wave vector component along the propagation direction, which means a larger resonant angle.This also proves that there is indeed a coupling effect between the guided mode and the SPR mode.The increase of FWHM may result from the weakening of the coupling effect.Moreover, we found that the phase change gradually tends to be flat with the increase in thickness (Figure 2d).This also illustrates that we can tune the thickness to control and adjust the coupling of two modes and realize a narrow high Q-value resonant peak accompanied by a phase mutation.Besides, we also investigated the effect from the thickness of metal layer (seeing the Support-ing Information).The influence of metal thickness on Q-value is relatively small, mainly affecting the depth of the resonance peak.
Following the simulation, we fabricated the chip with a waveguide layer of SiO 2 (h g = 480 nm) and a metal layer of Ag (h t = 48 nm) on a K9 glass.Through actual measurement, we obtained a very narrow resonant peak with FWHM ≈0.2°and Q-value ≈314 (see Supporting Information for calculation of Q-value), as shown in Figure 2e.Two modes coupling shows a narrow peak with angular FWHM of 0.2°, which is one order smaller than the measured angular spectrum of bare gold-based SPR (2.2°) and 6 times smaller than silver-based SPR (1.2°) (the insert in Figure 2e).From Figure 2f, we also found that owning to the mode coupling, the mode field at the resonant peak distributed near the sensing interface.This mode distribution avoids the propagation loss of metal surfaces, leading to the narrow resonance.On the other hand, this makes the sensor more sensitive to the external environment.The transformation of mode field distribution also proves the coupling effect between the two modes.

The Method and System for Phase Extracting
We fabricated the microfluidic sensing array chip (Figure 1d) with a photolithography patterning process and lift-off process (Figure S2, Supporting Information), as shown in Figure 1c,d for high-throughput measurement.A phase-polarization modulation scheme and a corresponding demodulation method were also adopted to extract and calculate the phase.In Figure 3a, we built the optical system for the phase extraction based on the two-modes coupling effect.In the system, the linear analyzer device is fixed on a programable rotary stage, which is controlled to modulate the polarization of reflected light in equal rotation steps.This kind of modulation transfer the light parameters to the modulation state (Figure 3a).A pair-prism module is used for this system, which can ensure that the direction of reflected light keeps consistent with the incident direction.Thus, whilst adjusting the optical system (such as rotating the incident angle), it is not necessary to perform big moves to the subsequent device.
The equal step change of the analyzer angle is achieved through program-controlling of the electric rotator.The modulation of the polarization state of reflected light appears as a modulation of the grayscale value of the CCD image (Figure 3b), which is essentially a kind of sinusoidal modulation.By fitting the polarization modulation curve, we obtained the correlation coefficient of the modulation curve, thereby demodulating the information of the polarization state.Since S-polarized light has no wave vector component along the propagation direction, the SPR effect cannot be excited.There was no other additional phase change for S-polarized light.Thus, by demodulating the polarization modulation curve, the phase information of reflected light could be also obtained.In the whole optical system, P-polarized light, and S-polarized light were completely in a common optical path.
To study the deviation between the extracted phase retardation and the real phase retardation, a quarter-wave plate was introduced into the system for verification.It was ensured that no other factors other than the quarter-wave plate affect the phase retardation of light.By fitting the modulation curve (Figure 3c-i,ii), TM + E 2 TM = 2).This means that we may achieve the dualparameters sensing including the amplitude and phase at the same time.The specific details of calculation and demodulation method are mentioned in the Supporting Information.For the phase retardation introduced by the /4 wave plate with a center wavelength of 532 nm, the demodulation matrix was calculated as [−01 593, −0.3 202, and 54.5 168] and the measured phase retardation was ≈69.11°and for the /4 wave plate with a center wavelength of 808 nm, the measured phase retardation is ≈108.92°.The measured value was very close to the real phase retardation (69°and 109°), which verified that this method can accurately extract the phase retardation between P-polarized light and S-polarized light.Moreover, as shown in Figure 3c-iii, we also performed polarization modulation on the linear-polarized light and demodulated its phase retardation.The coefficients of the modulation curve →0, →1, and the calculated phase retardation approached 0. We found that the demodulated phase retardation might have an imaginary part when it comes to the linear-polarized state.This might result from the existence of the imaging noise that might bring about fitting errors.

The Performance and Demonstration of Phase Sensing
The adoption of the pair-prism configuration (Figure 3a) can adjust the angle of incident light without affecting the setting of subsequent optical elements.The beam undergoes two total reflections in two parallel prism surfaces (see the insert of Figure 4a).After twice deflection, it exits along the direction parallel to the incident light.This is beneficial for us to observe the coupling effect in real-time through the imaging CCD during the adjustment process, as we need to scan the rotating angle of the prism stage (equivalent to adjusting the incident angle) to find the optimal resonant angle of the two-modes coupling.During the angle scanning (through rotating the platform together with the pair-prism), we found that the reflected light intensity underwent a process of first decreasing and then increasing (Figure 4a).The altering of reflected light intensity stands for the change of coupling degree.The incident angle with the strongest resonant effect can hence be determined during the process without the movement of other devices.
The determination of the optimal configuration was based on the lowest gray value of the CCD image during the angle scanning, to find the best coupling condition, as shown in Figure 4a.The methanol solutions of different mass fractions (0%, 0.1%, 0.3%, 0.5%, 1.0%, and 3.0%) were prepared as the gradient sample solutions for this performance testing.The solvent was deionized water.0% mass fraction means pure deionized water without the methanol.The size of the microfluidic chamber was designed to fully cover the sensing area.The microfluidic layer (PDMS-based) was sealed to the sensing chip (silica substrate) with the oxygen plasma treating.These methanol solutions were sequentially injected into the microfluidic chamber with the syringe and silicon tube, as shown in Figure 1a,d.The phase-polarization modulation and detection were carried out.As shown in Figure 4b, we obtained the phase-polarization modulation curves of the deionized water and the different concentration methanol solutions.The difference among the phase-polarization modulation curves indicates the change of the differential phase between the orthogonally polarized components, i.e., phase retardation.On the other hand, the change in the amplitude of the P-polarized component also occurs, which means a decrease in coupling efficiency.Furthermore, in Figure 4b, the phase polarization modulation curves of the solutions with different concentrations have the same grayscale value at 90°modulation state (i.e., the relative angle of the polarizer).This results from that the grayscale value at 90°m odulation state is contributed by the S-polarization component of the signal light.Since the two-modes coupling effect cannot be excited by the S-polarized component, it is not affected by the change of refractive index and stays at the same intensity.
According to the phase polarization modulation curve shown in Figure 4b, the differential phase for methanol solution with different concentrations can be demodulated and calculated.We found that the differential phase varieties sharply and satisfies linearity approximately at low concentrations (approximately within the range of 0% to 0.5% mass fraction) and gradually and continuously slow down with the increase of concentration (Figure 4c).The refractive index n meets a good linear relationship with the mass percentage c p (R 2 = 0.99818), as shown in the inset of Figure 4c.The refractive index of methanol is fitted based on the data in the Handbook of Chemistry and Physics.Through linear fitting and extrapolating, we obtained that the slope of refractive index n relative to mass percentage c p is ≈0.00023RIU.Therefore, we concluded that the refractive index difference of methanol solution with 0.1% mass fraction (c p = 0.1) relative to the deionized water is 0.000023 RIU (2.3 × 10 −5 RIU).The change index of phase response with respect to refractive index can be calculated as 1.1 × 10 5 degree RIU −1 .As the concentration of methanol solution increased, the refractive index changed accordingly, resulting in a further mismatch of resonant coupling conditions.With the increase of concentration, the change of differential phase tended to be smooth, and the sensitivity also decreased accordingly, as shown in Figure 4d.The sensitivity is ≈3-4 times that of the conventional phase interrogation based SPR sensor. [35,36]Meanwhile, we obtained that the angular sensitivity of the resonant peak is ≈60°RIU −1 (Supporting Information).Therefore, we inferred that the figure of merit (FOM) is ≈300 RIU −1 (the insert diagram of Figure 4d).The calculation of FOM is in Supporting Information.[39] Meanwhile, for the dual-parameters interrogation, we also obtained the intensity parameter from the demodulation coefficient matrix (coefficient ), as shown in the top insert diagram in Figure 4c.This intensity interrogation showed a considerable sensitivity of 7.2 × 10 4 a.u.RIU −1 , owing to the high Q feature of this sensor.Intensity interrogation showed a relatively better linear property and wider measurement range, but a relatively lower sensitivity.The additional intensity interrogation can expand the application range of ordinary, single phasetype sensors.This dual-parameters interrogation sensing architecture may provide the potential for the complementary and multi-sensing fusion in some application scenarios.

The Application and Operation for Alpha-Synuclein Detecting
The detection of alpha-synuclein is conducive to our further study of Parkinson's disease and has considerable significance for the diagnosis and treatment.The sensor developed in this paper is then applied to the detection of alpha-synuclein.The anti-alphasynuclein protein cannot bond itself directly to the sensing surface.We first modified a layer of the conjugate molecules on the sensing interface, which could be combined with the sensing surface and anchored on the surface.The conjugate molecules could also bind to the anti-alpha-synuclein protein.Thus, the anti-alpha-synuclein protein was anchored to the sensing interface indirectly.Here, polydopamine was selected as the conjugate molecule to immobilize the antibody as it has strong adhesion and can be anchored to the surface of almost any material.42] We prepared a 1 mg/mL dopamine solution (Tris buffer, pH 8.5) to modify the sensing interface and realize surface functionalization (Figure 5a).Under an alkaline environment, dopamine will self-polymerize into polydopamine.After functionalizing the sensing interface, the anti-alpha-synuclein protein was incubated on the surface and Bovine Serum Albumin (BSA) protein was used for blocking the remaining binding sites (Figure 5b).After the anti-alpha-synuclein protein was anchored to the sensing surface, we prepared and injected the alpha-synuclein solution with a known concentration ranging from 1 ng mL −1 to 1 μg mL −1 .The differential phase information of reflected light, as shown in Figure 5c, was collected and demodulated.Finally, we obtained the calibration response curve (the insert in Figure 5c) of the differential phase concerning the concentration of alphasynuclein which also showed the information on response sensitivity.Referring to the fitting curve, we speculated that the detection limit of the proposed sensor actuator for alpha-synuclein was ≈300 pg mL −1 level.Meanwhile, to verify the specificity of the SPR biosensor [43] Intensity-interrogation 2.71 × 10 4 a.u.RIU −1 ∖ 146 RIU −1 Single Enhanced SPR biosensor [39] Wavelength-interrogation 1015 nm RIU −1 136 108 RIU −1 Single Metasurface biosensor [29] Wavelength-interrogation 986 nm RIU −1 520 32.7 RIU −1 Single Enhanced SPR biosensor [10] Phase-interrogation 1.8 × 10 4 degree RIU −1 ≈30 ∖ Single Metasurface biosensor [44] Phase-interrogation 3.0 × 10 4 degree RIU −1 ≈37 ∖ Single This work Dual parameters-interrogation 1.1 × 10 5 degree RIU −1 (phase) 7.2 × 10 4 a.u.RIU −1 (intensity) 314 300 RIU −1 Dual Parameters detection, we injected 1 μg mL −1 BSA solution (the solvent was Phosphate Buffered Saline (PBS) buffer) to compare.We found that the differential phase response of BSA solution was very low.The response value was ≈5-6% of the target molecule under the same concentration, which indicated that the proposed sensor had a good specificity, as shown in Figure 5d.In conclusion, we applied the proposed enhanced refractive index sensor, where the waveguide mode coupling with the SPR mode and induces the phase mutation, to the detection of alpha-synuclein protein.This proved that the sensor and the system had an appreciable potential and a considerable reference value in the disease diagnosis.

Conclusion
In short, we considered that the coupling of guided mode and SPR mode supported a high-Q resonant peak, and the FWHM of the resonant peak was very narrow and accompanied by a drastic phase change, addressing the inherent loss issue of plasmonic metals.A simple configuration of a silica layer on the metal surface was used to impel the guided mode to couple with the SPR mode.By exciting the SPR mode with the evanescent wave of total reflection, the guided mode supported by the dielectric was simultaneously excited to couple with the SPR mode.
High-Q resonant peak is beneficial for both intensity and phase interrogation.We proposed a phase extraction method for the two-modes coupling based on a phase-polarization modulation and a corresponding demodulation algorithm.This scheme simplified the optical path of the sensing system without introducing any other bulky modulation devices.Combining with a pair-prism module, this scheme can promote the integration of the sensing system.In the sensing performance test, we obtained that the phase sensing sensitivity is ≈1.1 × 10 5 degree RIU −1 , which is >3-5 times the standard phase interrogation SPR sensor and the FOM value can reach 300 RIU −1 .Meanwhile, for the dualparameters interrogation, the intensity interrogation showed a considerable sensitivity of 7.2 × 10 4 a.u.RIU −1 with a larger detection range than phase interrogation.To compare the detection performance with other type of biosensor devices and methods, we also present the comparison of some key indicators (such as FOM or sensitivity) which are accepted metric for characterizing the performance of a refractive index sensor, as shown in Table 1.
From the table, we can see that the various performance indicators of our sensor basically have some advantages, especially the higher sensitivity of phase sensing and higher FOM.In addition, the dual-parameters interrogation also makes this sensor have a wider range of applications and higher potential for applica-tion.Finally, we applied the proposed sensing chip to the detection of alpha-synuclein protein.The limit of detection for alphasynuclein protein was estimated to be as low as 300 pg mL −1 level with a good specificity.The proposed sensing platform showed excellent sensing performance and might give full play in the field of sensing, disease diagnosis, and biochemical research in the future.
In the later study, one may use other materials (such as a higher refractive index or more biocompatible media) to realize the two-modes coupling between the guided mode and the SPR mode.Because a higher refractive index dielectric can reduce the waveguide thickness to realize the coupling condition, which means that it may be more affected by the external environment.Furthermore, we may replace the depositing method with a simpler spin coating method for active material polymers, which is conducive to high-throughput sensing and mass production.

Experimental Section
Fabrication: Titanium and silver were deposited by electron beam evaporation (Denton Vacuum, Explorer) with thicknesses of 1 and 48 nm, respectively.The silica layer was deposited on the surface by the Plasma Enhanced Chemical Vapor Deposition (PECVD) (STS, MESC MULTIPLEN CND) with a thickness of ≈480 nm.Its cross-section SEM (Raith GmbH, RAITH 150) is shown in Figure 1b.The stylus profiler (Bruker, DektakXT) was also used to characterize the thickness.The surface roughness of the deposited silica waveguide layer by PECVD was characterized by the atomic force microscopy (AFM) (Being Nano-Instruments, CSPM5500), and its roughness Ra was <3 nm (Figure S7, Supporting Information).The refractive index of the silica layer was measured by an ellipsometer (Eoptics, SE-VM), which is shown in Figure S6 (Supporting Information).The real part of the refractive index near 671 nm is ≈1.4612, and the imaginary part was near zero.
Measurement: The beam expanded module uses a 40× microscope objective lens and a shaping lens to expand and amplify the laser beam.The setting of the aperture diaphragm could ensure that the laser beam had no stray spot interference and improve the intensity uniformity of the expanded beam.The setting of an adjustable attenuator could make it easy to adjust the incident light power and avoid overexposure of the CCD image.The step size of the polarization modulation was selected at 15°.The motorized rotation stage was controlled to make the analyzer perform modulation in the range of 0°to 180°.The imaging lens and CCD were controlled for image acquisition.The zero-order /4 wave plate with the center wavelength of 532 nm (Thorlabs, WPQ05M05-532) and 808 nm (Thorlabs, WPQ05M-808) was selected to introduce the different phase retardation and to verify the correction of the phase extraction method.The relationship between the phase retardation and wavelength is shown in Figures S3 and S4 (Supporting Information).

Figure 1 .
Figure 1.Guided mode induces phase mutation for SPR sensor of enhanced sensitivity.a) The schematic of the SPR sensor enhanced by guided mode coupling.From top to bottom are the microfluidic layer, the silica waveguide layer, the metal layer, the substrate, and the prism, respectively.The insert is the process of identifying the biomolecule target.b) The cross-section of the fabricated structure with the silica, silver, and substrate layer.The scale bar is 200 nm.c) The actual photo of the sensing chip.The scale bar is 1 cm.d) Liquid samples pass into and out of the sensing module combined with the microfluidic layer and the coupling prism.The scale bar is 1 cm.

Figure 2 .
Figure 2. The theoretical analysis and simulation of the two-modes coupling.a) The resonant angular spectrum of the SPR with gold, silver, and guide-SRP coupling mode with a silica layer.b) The curve showing how the resonant phase changes with the refractive index.c) The resonant angular spectra with an increasing thickness of the silica layer.The thickness increases ranging from 480-720 nm, from left to right in the picture.d) The resonant phase changes with refractive index with different silica thicknesses.From top to bottom, the thickness increases from 480 to 720 nm.e) The measured resonant angular spectrum of two-modes coupling.The FWHM of the narrow resonant peak is ≈0.2°.The insert diagram is the measured angular spectrum of silver and gold.The thicknesses of the silver and gold for SPR modes in (a,b) are both 50 nm.The silver layer and silica layer for two-modes coupling are 48 and 480 nm thick, respectively.f) The mode field distribution for the two-modes coupling.The dotted line represents the boundary between layers.

Figure 3 .
Figure 3.The phase extraction of the present two-mode coupling sensor.a) The schematic of the system for phase extraction incorporated with the imaging technology.b) The actual image and corresponding measured gray value by phase-polarization modulation method.The line represents the fitting curve.c) The measured phase-polarization modulation results for two wave plates: i) with a 109°phase retardation; ii) with a 69°phase retardation; iii) The measured results for the linear-polarized input (0°phase retardation).Here, , , and  are the coefficients of fitting term.

Figure 4 .
Figure 4.The performance for refractive index sensing of resonant phase.a) The angle adjustment to match the resonant coupling condition based on the pair-prism module.b) The phase-polarization modulation curves of the solution with different concentrations.c stands for the mass fraction.c) The response curve of differential phase as the concentration of methanol solution varies.The bottom insert diagram is the relationship between the refractive index and the concentration of the solution.The top insert diagram is the  response curve changing with the concentration of methanol, which represents the amplitude response curve.d) The relationship between the slope of differential phase and the concentration of the solution.The insert diagram presents the calculated Q-value and FOM.

Figure 5 .
Figure 5.The detection of alpha-synuclein protein based on the proposed SPR sensor and phase extracting system.a) The modification process of the polydopamine conjugate molecule.The insert is the schematic of the molecular detection mechanism.b) The anti-alpha-synuclein modification process and the blocking process of binding sites.c) The differential phase response of alpha-synuclein protein with different concentrations.The insert is the responding curve as the concentration varied.d) The comparison of specific binding with BSA protein.The insert is the binding process of alpha-synuclein and BSA (both at the concentration of 1 μg mL −1 ) protein solution.

Table 1 .
The comparison of detection performance with other biosensor devices and methods.