Optical sensors with a high figure of merit (FOM) for refractive index measurement can substantially enhance detection performance. For guided mode resonance (GMR) sensors, previous works mainly focused on the sensitivity enhancement rather than FOM optimization; therefore, the state-of-the-art FOM is limited within the range of 100. To address this, we propose a low-index, ultraviolet-curable resin (n = 1.344) to form a simple, stable, symmetric, GMR sensor, with enhanced sensitivity, narrowed resonant linewidth, and substantially improved FOM, in aqueous media. The influence of structural parameters was systematically investigated, and optimized FOM values as high as tens of thousands were obtained using numerical calculation. Using low-cost, nanoimprinting technology, we experimentally demonstrated a spectral linewidth as narrow as 56 pm, a bulk refractive index sensitivity of 233.35 nm / RIU, and a low detection limit 1.93 × 10−6, resulting in a FOM value up to 4200 (48 times typical GMR sensors). The proposed symmetric GMR sensor exhibits great potential in a variety of applications, including label-free biosensing, bio-imaging, and optical filters.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
The guided-mode resonance (GMR) effect, by virtue of simple structure, easy detection schemes, high efficiency (up to 99%) and narrow linewidth (less than 1 nm) , has exerted considerable influence on many applications, such as polarizers , tunable filters [3–5], color filters [6,7], wave plates , biosensors [9–16] and bio-imaging [17–20]. A typical GMR device consists of periodic gratings and a high-refractive index waveguide layer . When light illuminates the structure, the incident light, which satisfies certain phase-matched conditions, can be coupled into the waveguide mode. Due to the periodic grating modulation, part of the light will out-couple, and will interfere with reflected light constructively, and transmitted light destructively, resulting in a sharp and high efficiency resonance spectrum, at a particular wavelength and polarization . The resonant light is highly sensitive to the change of surrounding refractive index, and furthermore, with the help of low-cost nanoimprinting technology, the structure area can be fabricated up to square centimeter level, making it available for high-throughput screening . This combination of properties suggests that GMR devices could be promising candidates for label-free biosensing.
Over the last decade, many research groups have worked on improving GMR sensor sensitivity, proposing several effective strategies. Lin et al. proposed a metal layer-assisted, GMR (MaGMR) device, that could provide stronger evanescent energy to the sensing area, —and showed that their MaGMR sensor could achieve 376.78 nm / refractive index unit (RIU) experimentally, which represented a 265% sensitivity enhancement over the basic structure . Ku et al. utilized different sputtering parameters to create a low-index cavity layer at the interface between the waveguide layer and the substrate. The sensitivity of their modified GMR sensor was 181.9 nm / RIU, which was a 2.2-fold improvement over sensors without a cavity layer . Wan et al. presented porous SiO2 with an ultralow refractive index (n = 1.09) as the substrate for a GMR sensor, —which relocated its resonant modes to reside in the sensing medium, —and in doing so achieved a four-fold increase (502 nm / RIU) over a typical GMR sensor .
Although sensitivity is the most common indicator used to gauge sensor performance, for optical resonance sensors, figure of merit (FOM), which combines the ratio of simultaneous availability of sensitivity (S) with the resolution of small surrounding changes (FOM = S / Γ, where Γ is the full width at half maximum (FWHM) of the resonant position), is a more comprehensive indicator to evaluate sensing performance [27–30]. In reality, sensors with high FOM values exhibit substantially enhanced detection ability towards the lower limits of detection.
The literature on GMR sensors has focused on sensitivity enhancement rather than FOM optimization, with the result that the FOM for GMR sensors is still commonly limited to a range of ∼ 100 [23–26,31–34]. Recently, while few works have investigated this important indicator [35–37], they have been able to show in computer simulation that FOM for GMR sensors can be improved very significantly, up to 12000, due to extremely narrow resonant linewidth (∼ 0.02 nm) and high efficiency . Experimental demonstration of high FOM for GMR sensors remains elusive however, and it should be noted that achievement of such a narrow resonant linewidth, which is easily predicted by infinite plane wave models, is a challenging task in reality, —especially for transverse magnetic (TM) polarization, —as the resonant signal may be hard to distinguish from background noise . Moreover, the signal is also sensitive to the degree of fidelity between the design and practical fabrication of GMR sensors.
A typical GMR sensor usually has an asymmetric structure, since the substrate refractive index is higher than that of the surrounding medium. A suspended symmetric structure (without a substrate layer) can enhance GMR sensor performance (high sensitivity, narrow linewidth, and low detection limits) thanks to the enlarged biosensing volume and electric field energy that is normally localized in the waveguide layer [39,40]. Such a suspended symmetric structure is difficult to fabricate, and is very fragile in practice . This makes it very important to develop a symmetric GMR sensor with a simple structure, but with high sensitivity and spectral resolution.
In this work, an ultraviolet curable resin (UVCR) with a refractive index similar to water (n = 1.344), was used to fabricate a symmetric structure in an aqueous medium. This compact structure was more stable and solid compared with a suspended structure, and could be achieved using simple and low-cost nanoimprinting technology, —thus paving the way for achieving the high-throughput industrial mass production for such a sensor.
Importantly, the high performance of this symmetric structure has been demonstrated experimentally, with the TM resonant signal successfully detected with a narrow linewidth of 56 pm. The FOM was as high as 4200, or 48 times the values of typical GMR sensors previously reported, and comparable to the reported values achieved by photonic crystal (PhC) cavity sensors , microring sensors [42,43], and fiber Bragg grating (FBG) sensors . Besides, low detection limits (DL) of 1.93 × 10−6 was demonstrated. Such a high performance GMR sensor shows great potential in biosensing, and in other applications, including optical filters and label-free bio-imaging.
2. Structure optimization
The proposed GMR sensor is shown in Fig. 1. A low refractive index UVCR (nl) was used for the substrate and for gratings with a period of Λ, depth of groove dg, and filling factor f. The grating surface was covered with a high refractive index material (nh), with the height of dwg, and has been referred to here as the waveguide and sensing layer. For more realistic consideration , the thin, high-index layer also covered the side walls of the grating ridges (ds), with the ratio of dwg / ds = 1.5.
We used COMSOL Multiphysics 5.2a (COMSOL Inc., Stockholm, Sweden), which is based on the finite element analysis (FEA) method , to calculate the field distributions between symmetric and asymmetric cases. Rigorous coupled-wave analysis (RCWA), which can obtain results similar to FEA, was also used to simulate the GMR effect .
TM polarized light (with the magnetic field Hz perpendicular to the plane of incidence (x–y plane)), which has higher sensitivity and a narrower resonance linewidth than transverse electric (TE) polarization, was used to generate the electromagnetic mode . For our simulations, Λ was set at 505 nm, dg at 150 nm, nh at 2.1, dwg at 50 nm, f at 0.5, and the refractive index of the cover medium (nc) was set at 1.333. The index, nl, was set to 1.45 when a relatively high-index substrate compared to water was required for the asymmetric structure, and to 1.344 when a similar refractive index to water was required for the symmetric structure substrate medium. In the calculation, the left and right boundaries of the computational domain are set as periodic boundaries and the distance between the two boundaries is the exact value of one grating period. The top and bottom boundaries of the computational domain are set as ports with a distance of around 4 µm in between.
In asymmetric structure, the electric field mainly resides in the substrate regions of the structure and only ≈ 200 nm depth of evanescent field extends into the cover medium (Fig. 2(a)), —and resonant linewidth and sensitivity are limited because of that. As observed in Figs. 2(b) and 2(c), the resonant linewidth in the asymmetric structure was 0.75 nm, the sensitivity (S =Δλ / Δnc, where Δλ is the shift of resonant peak wavelength divided by Δnc, which is the change of surrounding refractive index,) was 132.96 nm / RIU, and so the corresponding FOM was 177. On the other hand, the electric field in the symmetric structure was normally localized in the waveguide layer, and the more evanescent field (≈ 400 nm depth) tended towards the sensing region (Fig. 2(d)). As observed in Figs. 2(e) and 2(f), although the resonant linewidth did not have a distinctly narrowing appearance (Γ = 0.66 nm in the symmetric case), sensitivity was clearly enhanced, to 204.56 nm / RIU, so the corresponding FOM achieved was 310. Moreover, to give a clear insight into the influence of nl, the sensitivity, resonant linewidth, and FOM as a function of nl for the TM polarization is shown in Fig. 2(g). Lower nl has a higher sensitivity because the resonant mode resides mainly in the sensing medium instead of the substrate medium, but the broadened linewidth at lower nl compromises the FOM value. Therefore, nl around 1.33 is the best trade-off choice.
To facilitate comprehensive discussion on the proposed symmetric structure, and to optimize its performance, all parametric values, including Λ, nh, dwg, dg, and f, were investigated in detail. In addition, the performance of asymmetric structure was also shown as the control group for comparison.
Λ is the primary parameter, which influences GMR sensor resonant peak wavelength and sensitivity. The resonant wavelength can change significantly if Λ varies in a large range. Relevant literature reveals that a higher Λ leads to higher peak wavelength and higher sensitivity [34,40], while it is unclear whether Λ will influence linewidth or the corresponding FOM. To address this issue, different values for Λ were chosen to calculate sensor performance, as shown in Figs. 3(a) and 3(b). Sensitivity has a linear relationship with Λ, and the resonant linewidth initially increased, and then decreased in symmetric structure (see Fig. 3(a)). Overall FOM (violet marked symbols) increased as Λ increased, in Fig. 3(a), and considering these results, if the aim was to achieve a high performance GMR sensor, a higher Λ was an effective method, using the proposed structure. Contrast to the symmetric structure, sensitivity does not show a linear relationship with Λ in asymmetric structure (see Fig. 3(b)). Besides, owing to that the dwg is only 50 nm in our calculation for the asymmetric structure, the GMR effect will disappear once the Λ is larger than 700 nm, which makes the GMR effect locate near the cutoff region .
To fit our subsequent experiments, a 505-nm Λ was used when we studied the influence of other parameters. Variables nh and dwg are also important factors with respect to the performance of GMR sensors, especially in sensitivity , and while previous works have ignored the effect of nh on FOM, here, we varied nh to identify any FOM variation trend, with other parameters remaining as shown in Fig. 2(d). As shown in Fig. 3(c), decreased nh accompanied increased sensitivity and narrowed linewidth, as the FOM rose towards 22536 (at which point sensitivity was 233.22 nm / RIU, and linewidth was 10.35 pm) when the nh was set at 1.6 (see Fig. 3(c)). The narrow resonant linewidth was attributed to the relatively low contrast between the refractive indexes of the substrate and the waveguide layer, which resulted in longer guided mode lifetimes . In addition, lower nh would lead to greater electric field distribution extent in the sensing region, thus increasing GMR sensor sensitivity . These results indicated that a relatively low refractive index material was more suitable for achieving higher FOM in the symmetric structure. In contrast, with the increasing nh, the sensitivity of the asymmetric structure shows an opposite trend compared with that of the symmetric structure, as shown in Fig. 3(d), since the phase shift here plays a more dominating role .
To analyze performance change due to dwg variation, nh was fixed at 1.7 for these calculations, and the other parameters were kept as shown in Fig. 3(c). The results, as shown in Fig. 3(e), indicated that sensitivity tended higher and linewidth decreased, with dwg thickness reduction; hence, as dwg decreased, a higher FOM was obtained, together with higher sensitivity and narrower resonant linewidth. This result was slightly different with the asymmetric GMR structure, where a most sensitive dwg region exists, as was shown in Fig. 3(f). This may be attributed to a phase difference that was not apparent in the symmetric structure considered here .
Variables dg and f have been widely studied in GMR filters. Generally, linewidth has been found to be either directly proportional to dg in “grating-waveguide” GMR filters , or has shown peaks and nulls in “waveguide-grating” GMR filters , while f controls the linewidth in a sinusoidal form .
On the other hand, these two parameters have seldom been investigated for GMR sensors. For symmetric structure, sensitivity was found to be inversely proportional to dg (see Fig. 3(g)), and as dg increased, linewidth initially increased, then decreased, until an extremely narrow resonant linewidth suddenly occurred, at dg = 250 nm, and then increased again. This phenomenon was attributed to the orthogonality of the bound mode and the radiation mode decreasing the overall magnitude of the coupling loss . Thanks to the extremely narrow linewidth, the highest FOM number was obtained at dg = 250 nm, although sensitivity was modest at this position. If higher sensitivity—together with narrow linewidth—were required, a thinner dg would answer. Figure 3(h) shows the changes for asymmetric cases. It can be observed that the sensitivity increasing as the dg increases, which is different with the symmetric case, but, the dependences of linewidth to dg have similar trend for the symmetric and asymmetric structures.
Regarding the influence of f, it was seen to be contributing to sensitivity control in a “U” shape, and to linewidth control in an almost sinusoidal form (Fig. 3(i)). The overall FOM clearly increased (accompanied by high sensitivity and narrow linewidth) when f was larger, as shown in Fig. 3(i). In asymmetric structure, f controls both the sensitivity and linewidth in a sinusoidal form, as shown in Fig. 3(j).
It should be noted that these optimized FOM results in symmetric structure (easily achieving values in the tens of thousands) were greater than those in asymmetric structure (generally limited to a thousand), as shown in Fig. 3. Besides, these optimized results are way better than experimental values reported to date [23,31–34], and even higher than the results achieved in simulations [35–37]. This was due to the nearly symmetric structure in an aqueous environment allowing the electrical field energy to be located uniformly in the cover and substrate medium, giving rise to higher sensitive and narrower linewidth. According to the simulation results, the main contributors to higher FOM (accompanied with higher sensitivity and narrow linewidth) were higher Λ, or lower nh, or thinner dwg, or smaller dg, or larger f in this symmetric structure. Depending on the different experimental conditions applied, one or more of these parameters could be selected for application to the optimized sensor structure, and then deployed in its fabrication.
3. Device fabrication
Nanoimprinting technology is used widely to fabricate grating structures, thanks to its low cost and convenience . A silicon master contains a 10 × 10 mm, 1-dimension, linear grating area, with a duty cycle (f) of 0.5, groove depth (dg) of 150 nm, and a period (Λ) of 505 nm, —and includes a thin layer of perfluorooctyltrichlorosilane deposited onto the master for anti-stiction.
For our device, the glass substrate was ultrasonically rinsed in acetone and then isopropyl alcohol, for 15 min respectively. After being blown dry in a nitrogen stream, the substrate was baked at 200 °C, for 15 min, and then rapidly annealed by air. Oxygen plasma was then applied, for 15 min, to oxidize the glass substrate.
As depicted in Fig. 4(a), a small amount of SPC-347 UVCR (FOSPIA EFiRON, Ansan, Korea) was dropped onto both the substrate and the grating area of the master. Then the master was joined to the glass substrate, for 2 min, as the SPC-347 UVCR spread out due to the combined effect of the weight of the glass and capillary forces. Ultra-violet (UV) light from a UV lamp was applied for two min to solidify the SPC-347 UVCR. The glass substrate was then separated from the silicon master using a blade, and the thin SPC-347 film, which contained the periodic gratings, was left on the glass substrate. Finally, UV light was applied for 2 h, as a post-separation treatment, in order to ensure the solidity and stability of the UVCR.
A commercial pure titanium dioxide (TiO2) target was used for the preparation of TiO2 thin films by radio-frequency (RF) magnetron sputtering. The sputtering was performed while keeping the RF power at 108 W, using a gas mixture of Ar (0.5 Pa) and O2 (2 sccm) atmosphere at room temperature. The result was a structure using low refractive index (1.344) UVCR, —which has a refractive index similar to that of water, —for the substrate and gratings of the sensor, and high-index (TiO2) sputtering above the gratings as the waveguide layer, to support the GMR effect.
Figure 4(b) is a scanning electron microscope (SEM) image of the grating structure, with a Λ of 505 nm, which verified the success of nanoimprinting the SPC-347 UVCR. As observed in Fig. 4(c), the gaps between grating ridges became smaller after TiO2 film deposition, which verified that a thin layer had been added to both sides of the grating ridges, giving a result similar to that previously published . An image and a photo of a fabricated, 10 × 10 mm2 GMR sensor are shown in Figs. 4(d) and 4(e), respectively.
4. Experimental results
The experimental setup was designed as shown in Fig. 5. A supercontinuum source (470–2400 nm, item SC-5-FC, YSL Photonics, USA) was employed as the light source, and was collimated by a fiber collimator, focused with a lens and then polarized with a polarizer to illuminate the GMR device. TE polarization corresponds to light whose electric ﬁeld is parallel to the grating lines, while TM polarization corresponds to light whose electric ﬁeld is perpendicular to grating lines. The reflected light passed through a beam splitter into the spectrometer with a wavelength resolution of 34 pm (SpectraPro-2750, Acton Research, USA). The microfluidic channel consists of a microslide sandwiched between two metallic plates, namely the front metallic plate (with a hollow region to exposure the sample area) and back metallic plate (with the flow channels to inject the liquid). An elastic was inserted between the back metallic plate and the microslide which contains the sensor.
In previous theoretical discussions, a thinner waveguide layer depth, dwg, was considered useful for achieving higher FOM. We fabricated three different TiO2 thin film thicknesses for the waveguide (sensing) layer. Experimentally, the performance of the low index UVCR-based symmetric structure could be evaluated by immersion in different refractive index aqueous environments, corresponding to deionized (DI) water and several dilutions of dimethyl sulfoxide (DMSO).
Figures 6(a) and 6(b) show measured spectra for a 35-nm thickness, TiO2 thin films-deposited GMR sensor. Only TE-polarized mode is shown, as the TM-polarized mode was hard to identify (physical interpretation of the issue will be attempted in the Section 5). The concentrations of the DMSO solution in water were diluted to 0% (DI water), 2%, 4%, 6%, 8% and 10%, which corresponds to refractive indices ranging from 1.333 to 1.3477, and the peak position as a function of refractive index is also shown in Fig. 6(b). To further investigate the linear dynamic range, DMSO solutions were diluted to 15% and 20%, corresponding to the refractive index of 1.355 and 1.3624, respectively. As shown in Fig. 6(b), a wide linear detection range of 0.0294 RIU was achieved, in good agreement with recently reported GMR sensor , and the sensitivity of sensor was 234.72 nm / RIU. Meanwhile, the resonant linewidth was 1.46 ± 0.02 nm in DI water, which corresponded to a FOM of 161 ± 2 (for comparison, in simulation, S = 260.67 nm / RIU, FWHM = 1.17 nm, FOM = 190).
To understand the different effects of TM and TE polarization better, electric field distributions were calculated, as shown in Figs. 6(c) and 6(d). It can be clearly seen that the electric field was more concentrated for TM-polarized mode. In addition, the normalized field intensity across the y axis was plotted in Fig. 6(e), where the blue dashed line stands for the bottom interface of the high-index waveguide layer. In this structure, more energy was scattered in the substrate layer for TE-polarized mode, as a consequence of which energy was wasted in the substrate, while the TM-polarized mode scattering in the substrate region was less, giving rise to longer resonating lifetimes.
With the aim of high efficiency and narrow linewidth, spectra were measured for TM-polarized mode, 50-nm thickness, TiO2 thin films-deposited GMR sensors, —and the results can be seen in Figs. 7(a) and 7(b). Resonant spectra TM linewidth as narrow as 0.056 ± 0.007 nm (with the quality factor reaching up to 12000) were successfully detected, with sensitivity of 233.35 nm / RIU, which corresponded to a FOM of up to 4200 ± 600, —which was 48 times greater than the highest reported GMR sensor values. This experimental result agreed well with the simulation (S = 221.09 nm / RIU, FWHM = 0.038 nm, FOM = 5800), as shown in Fig. 3. In addition, it should be noted that the experimental FOM for our GMR sensors was even better than some other optical resonant sensors, such as PhC cavity sensors (FOM = 1100) , microring sensors (FOM = 1200) [42,43], and FBG (FOM = 2500) .
Figures 7(c) and 7(d) show the measured results for 70-nm thickness, TiO2 thin films-deposited GMR sensors. In this case, the linewidth was 0.45 ± 0.05 nm, with sensitivity of 215.36 nm / RIU, corresponding to a FOM value of 500 ± 60. Broadened linewidth and reduced sensitivity were observed here, compared with the 50-nm version. These experimental results coincided with the results calculated previously, showing that thicker dwg led to wider resonant linewidth and lower sensitivity. Besides, a wide linear detection range of 0.0294 RIU were achieved in both two cases, which is sufficient for typical bio-analytical applications.
The slight divergence between the simulation and experiment was probably due to fabrication deviation in making the grating and thin films, that is, grating and TiO2 thin film inclination was not accounted for in the model, and for this reason, the ratio of ridge (including grating ridge and TiO2 thin films on both sides of the grating ridge) to groove in our calculated model was slightly different to the actual scenario. Besides, in contrast with the simulation results shown in Figs. 2(b) and 2(e), the resonances are not smooth in the experimental demonstration, which have negative impact to the sensitivity and FOM. This phenomenon is to the Fabry-Perot (FP) interference caused by the glass substrate , which can be eliminated by coating anti-reflection films.
When the narrower resonant linewidth was attempted, the resonance signal started to be very sensitive to any loss mechanism, which in practice can make the actual reflectance efficiency much lower than unity. Thus, although narrow linewidth was easily achieved in stimulation, it was harder to obtain in reality, especially for TM polarization , so, in our experimentation, we needed to explore the optimized condition further. Figure 8(a) shows the normalized spectra for 35-nm, 50-nm, and 70-nm thickness deposited TiO2 thin films sensors, in DI water. Clearly, very low background noise was seen for 35-nm TE-polarized light, which indicated the reflected efficiency at the resonant position was very strong, as observed in Fig. 8(b). For the 50-nm case, the TM-polarized light had a distinct noise signal compared with the other two cases. When the noise signal was stronger, this ultra-narrow linewidth could not be clearly observed, as for the 35-nm case where only TE-polarized light was obtained. A thicker film (70 nm case) was helpful in eliminating strong background noise, and the measured linewidth was broadened, leading us to speculate that the 50-nm case might be a good point at which the resonant signal could be identified from the background noise, with extremely narrow linewidth and relatively high efficiency. For 50-nm case, the resonance peak wavelength shift from a set of 50 spectra were depicted in Fig. 8(c). We observed that an excellent stability of resonance peak wavelength (standard deviation σ = 0.15 pm) was achieved. This yields a detection limit (DL = 3σ / S) of 1.93 × 10−6 for proposed symmetric GMR sensor, which is also superior to existing reported values [12–13,23].
Having demonstrated the performance of symmetric GMR sensors in an aqueous medium based on SPC-347 UVCR, we have summarized the linewidth, refractive index sensitivities, and FOM values for asymmetric GMR sensors and other varieties of optical sensor technology from the literature (Table 1). The FOM of 4200 ± 600 achieved herein was much higher than the values (within the range of 100) previously reported for GMR sensors, and was even superior to some of the reported values using PhC cavity, microring, and fiber Bragg grating (FBG) sensor technologies. It should also be noted that the simulation results for the symmetric structure exceeded calculations for the asymmetric structure [35–37]. Furthermore, compared to asymmetric GMR sensors and other optical sensor varieties, this symmetric sensor has lower detection limit of 1.93 × 10−6. To outstanding the performance of proposed symmetric GMR sensor, we summarized the sensitivity, FOM and DL coming from asymmetric GMR sensors [31,38] and other optical sensors [41–43], as shown in Fig. 8(d). The symmetric structure should therefore be considered superior to the asymmetric structure, in both theory and experimental practice. However, it should be noted that the reduction of signal-to-noise ratio (SNR) can become a major problem for real sensors. For instance, the SNR reduction will increase the value of σ . Therefore, the practical FOM should consider the SNR. For example, the practical FOM can be defined using the following expression FOM* = FOM · (1 – NMBR), where NMBR is the maximum background reflectance normalized to the peak of resonance signal near the resonance position. A lower NMBR is required because a higher SNR allows an efficient detection of resonance shift and a precise analysis of sensing events.
In contrast to the suspended symmetric design, which has the drawbacks of additional fabrication steps and greater fragility, nanoimprinting has the advantages of low process complexity, cost effectiveness, production speed and good reproducibility, all of which support high-throughput, industrial mass production. This approach also has the advantage of making the device more stable and solid, compared with the suspended structure.
In addition, with ultra-narrow linewidths, this device could be a powerful tool for optical filters, particularly tunable optical filters, given its ability to respond to fine adjustment. Another positive is that its open-faced feature, that interfaces with liquid media, makes it a promising candidate for biosensing and bio-imaging.
In summary, in our study, a symmetric GMR sensor structure was created in an aqueous environment. The symmetric structure means that its electric field energy was uniformly located in the cover and substrate medium, giving rise to higher sensitivity and narrower linewidth. To optimize higher FOM values, its parametric properties were systematically investigated via simulation, and the results showed that ultra-high FOM could easily be achieved. Finally, the proposed sensor’s high performance was demonstrated experimentally, and the results agreed well with those achieved in simulations. Experimental results demonstrated a FOM value up to 4200 ± 600 (48 times larger than typical GMR sensors), which was accompanied by narrow linewidth (56 pm) and high sensitivity (233.35 nm / RIU), and a low detection limit of 1.93 × 10−6. This high performance GMR structure shows great potential in accurate biosensing, and for other applications, such as optical filters and label-free bio-imaging.
Shanghai Municipal Science and Technology Commission (16YF1400700, 18JC1411500); National Basic Research Program of China (973 Program) (2015CB352006); National Natural Science Foundation of China (11674062, 61327008, 61378080, 61505032, 61705039, 61705042); Funding Special Project of National Key Technology R&D Program of the Ministry of Science and Technology of China (2016YFC0201401).
The authors declare no conflicts of interest.
1. B. T. Cunningham, M. Zhang, Y. Zhuo, L. Kwon, and C. Race, “Recent Advances in Biosensing With Photonic Crystal Surfaces: A Review,” IEEE Sens. J. 16(10), 3349–3366 (2016). [CrossRef]
2. S. S. Wang and R. Magnusson, “Theory and Applications of Guided-Mode Resonance Filters,” Appl. Opt. 32(14), 2606–2613 (1993). [CrossRef]
3. D. W. Dobbs and B. T. Cunningham, “Optically tunable guided-mode resonance filter,” Appl. Opt. 45(28), 7286–7293 (2006). [CrossRef]
4. L. Y. Qian, D. W. Zhang, C. X. Tao, R. J. Hong, and S. L. Zhuang, “Tunable guided-mode resonant filter with wedged waveguide layer fabricated by masked ion beam etching,” Opt. Lett. 41(5), 982–985 (2016). [CrossRef]
5. C. L. Fang, B. Dai, Z. Li, A. Zahid, Q. Wang, B. Sheng, and D. W. Zhang, “Tunable guided-mode resonance filter with a gradient grating period fabricated by casting a stretched PDMS grating wedge,” Opt. Lett. 41(22), 5302–5305 (2016). [CrossRef]
6. D. B. Mazulquim, K. J. Lee, J. W. Yoon, L. V. Muniz, B. H. Borges, L. G. Neto, and R. Magnusson, “Efficient band-pass color filters enabled by resonant modes and plasmons near the Rayleigh anomaly,” Opt. Express 22(25), 30843–30851 (2014). [CrossRef]
7. M. J. Uddin, T. Khaleque, and R. Magnusson, “Guided-mode resonant polarization-controlled tunable color filters,” Opt. Express 22(10), 12307–12315 (2014). [CrossRef]
8. R. Magnusson, M. Shokooh-Saremi, and E. G. Johnson, “Guided-mode resonant wave plates,” Opt. Lett. 35(14), 2472–2474 (2010). [CrossRef]
9. B. Cunningham, P. Li, B. Lin, and J. Pepper, “Colorimetric resonant reflection as a direct biochemical assay technique,” Sens. Actuators, B 81(2-3), 316–328 (2002). [CrossRef]
10. B. Cunningham, B. Lin, J. Qiu, P. Li, J. Pepper, and B. Hugh, “A plastic colorimetric resonant optical biosensor for multiparallel detection of label-free biochemical interactions,” Sens. Actuators, B 85(3), 219–226 (2002). [CrossRef]
11. V. Canalejas-Tejero, A. López, R. Casquel, M. Holgado, and C. A. Barrios, “Sensitive metal layer-assisted guided-mode resonance SU8 nanopillar array for label-free optical biosensing,” Sens. Actuators, B 226, 204–210 (2016). [CrossRef]
12. G. J. Triggs, Y. Wang, C. P. Reardon, M. Fischer, G. J. O. Evans, and T. F. Krauss, “Chirped guided-mode resonance biosensor,” Optica 4(2), 229–234 (2017). [CrossRef]
13. Y. C. Lin, W. H. Hsieh, L. K. Chau, and G. E. Chang, “Intensity-detection-based guided-mode-resonance optofluidic biosensing system for rapid, low-cost, label-free detection,” Sens. Actuators, B 250, 659–666 (2017). [CrossRef]
14. G. Sancho-Fornes, M. Avella-Oliver, J. Carrascosa, E. Fernandez, E. M. Brun, and A. Maquieira, “Disk-based one-dimensional photonic crystal slabs for label-free immunosensing,” Biosens. Bioelectron. 126, 315–323 (2019). [CrossRef]
15. W. J. Kim, B. K. Kim, A. Kim, C. Huh, C. S. Ah, K. H. Kim, J. Hong, S. H. Park, S. Song, J. Song, and G. Y. Sung, “Response to cardiac markers in human serum analyzed by guided-mode resonance biosensor,” Anal. Chem. 82(23), 9686–9693 (2010). [CrossRef]
16. X. Wei and S. M. Weiss, “Guided mode biosensor based on grating coupled porous silicon waveguide,” Opt. Express 19(12), 11330–11339 (2011). [CrossRef]
17. W. L. Chen, K. D. Long, M. Lu, V. Chaudhery, H. Yu, J. S. Choi, J. Polans, Y. Zhuo, B. A. C. Harley, and B. T. Cunningham, “Photonic crystal enhanced microscopy for imaging of live cell adhesion,” Analyst 138(20), 5886–5894 (2013). [CrossRef]
18. W. L. Chen, K. D. Long, H. J. Yu, Y. F. Tan, J. S. Choi, B. A. Harley, and B. T. Cunningham, “Enhanced live cell imaging via photonic crystal enhanced fluorescence microscopy,” Analyst 139(22), 5954–5963 (2014). [CrossRef]
19. W. L. Chen, K. D. Long, J. Kurniawan, M. Hung, H. J. Yu, B. A. Harley, and B. T. Cunningham, “Planar Photonic Crystal Biosensor for Quantitative Label-Free Cell Attachment Microscopy,” Adv. Opt. Mater. 3(11), 1623–1632 (2015). [CrossRef]
20. Y. Zhuo, J. S. Choi, T. Marin, H. J. Yu, B. A. Harley, and B. T. Cunningham, “Quantitative analysis of focal adhesion dynamics using photonic resonator outcoupler microscopy (PROM),” Light: Sci. Appl. 7(1), 9 (2018). [CrossRef]
21. I. D. Block, N. Ganesh, M. Lu, and B. T. Cunningham, “A sensitivity model for predicting photonic crystal biosensor performance,” IEEE Sens. J. 8(3), 274–280 (2008). [CrossRef]
22. G. Quaranta, G. Basset, O. J. F. Martin, and B. Gallinet, “Recent Advances in Resonant Waveguide Gratings,” Laser Photonics Rev. 12(9), 1800017 (2018). [CrossRef]
23. G. Pitruzzello and T. F. Krauss, “Photonic crystal resonances for sensing and imaging,” J. Opt. 20(7), 073004 (2018). [CrossRef]
24. S. F. Lin, C. M. Wang, T. J. Ding, Y. L. Tsai, T. H. Yang, W. Y. Chen, and J. Y. Chang, “Sensitive metal layer assisted guided mode resonance biosensor with a spectrum inversed response and strong asymmetric resonance field distribution,” Opt. Express 20(13), 14584–14595 (2012). [CrossRef]
25. Y. F. Ku, H. Y. Li, W. H. Hsieh, L. K. Chau, and G. E. Chang, “Enhanced sensitivity in injection-molded guided-mode-resonance sensors via low-index cavity layers,” Opt. Express 23(11), 14850–14859 (2015). [CrossRef]
26. Y. H. Wan, N. A. Krueger, C. R. Ocier, P. Su, P. V. Braun, and B. T. Cunningham, “Resonant mode engineering of photonic crystal sensors clad with ultralow refractive index porous silicon dioxide,” Adv. Opt. Mater. 5(21), 1700605 (2017). [CrossRef]
27. M. Zhang, M. Lu, C. Ge, and B. T. Cunningham, “Plasmonic external cavity laser refractometric sensor,” Opt. Express 22(17), 20347–20357 (2014). [CrossRef]
28. M. Zhang, C. Ge, M. Lu, Z. X. Zhang, and B. T. Cunningham, “A self-referencing biosensor based upon a dual-mode external cavity laser,” Appl. Phys. Lett. 102(21), 213701 (2013). [CrossRef]
29. Q. L. Huang, J. Peh, P. J. Hergenrother, and B. T. Cunningham, “Porous photonic crystal external cavity laser biosensor,” Appl. Phys. Lett. 109(7), 071103 (2016). [CrossRef]
30. C. Ge, M. Lu, S. George, T. A. Flood, C. Wagner, J. Zheng, A. Pokhriyal, J. G. Eden, P. J. Hergenrother, and B. T. Cunningham, “External cavity laser biosensor,” Lab Chip 13(7), 1247–1256 (2013). [CrossRef]
31. S. Boonruang and W. S. Mohammed, “Multiwavelength guided mode resonance sensor array,” Appl. Phys. Express 8(9), 092004 (2015). [CrossRef]
32. I. D. Block, L. L. Chan, and B. T. Cunningham, “Photonic crystal optical biosensor incorporating structured low-index porous dielectric,” Sens. Actuators, B 120(1), 187–193 (2006). [CrossRef]
33. Y. K. Tu, M. Z. Tsai, I. C. Lee, H. Y. Hsu, and C. S. Huang, “Integration of a guided-mode resonance filter with microposts for in-cell protein detection,” Analyst 141(13), 4189–4195 (2016). [CrossRef]
34. M. Z. Tsai, C. T. Hsiung, Y. Chen, C. S. Huang, H. Y. Hsu, and P. Y. Hsieh, “Real-time CRP detection from whole blood using micropost-embedded microfluidic chip incorporated with label-free biosensor,” Analyst 143(2), 503–510 (2018). [CrossRef]
35. H. Lu, M. Huang, X. B. Kang, W. X. Liu, C. Dong, J. Zhang, S. Q. Xia, and X. Z. Zhang, “Improving the sensitivity of compound waveguide grating biosensor via modulated wavevector,” Appl. Phys. Express 11(8), 082202 (2018). [CrossRef]
36. G. L. Lan, S. Zhang, H. Zhang, Y. H. Zhu, L. Y. Qing, D. M. Li, J. P. Nong, W. Wang, L. Chen, and W. Wei, “High-performance refractive index sensor based on guided-mode resonance in all-dielectric nano-silt array,” Phys. Lett. A 383(13), 1478–1482 (2019). [CrossRef]
37. Y. Zhou, B. Wang, Z. Guo, and X. Wu, “Guided Mode Resonance Sensors with Optimized Figure of Merit,” Nanomaterials 9(6), 837 (2019). [CrossRef]
38. P. G. Hermannsson, K. T. Sorensen, C. Vannahme, C. L. C. Smith, J. J. Klein, M. M. Russew, G. Grutzner, and A. Kristensen, “All-polymer photonic crystal slab sensor,” Opt. Express 23(13), 16529–16539 (2015). [CrossRef]
39. M. El Beheiry, V. Liu, S. H. Fan, and O. Levi, “Sensitivity enhancement in photonic crystal slab biosensors,” Opt. Express 18(22), 22702–22714 (2010). [CrossRef]
40. S. F. Lin, C. M. Wang, Y. L. Tsai, T. J. Ding, T. H. Yang, W. Y. Chen, S. F. Yeh, and J. Y. Chang, “A model for fast predicting and optimizing the sensitivity of surface-relief guided mode resonance sensors,” Sens. Actuators, B 176, 1197–1203 (2013). [CrossRef]
41. Y. Liu, S. Wang, D. Zhao, W. Zhou, and Y. Sun, “High quality factor photonic crystal filter at k ≈0 and its application for refractive index sensing,” Opt. Express 25(9), 10536–10545 (2017). [CrossRef]
42. C. Chung-Yen, W. Fung, and L. J. Guo, “Polymer microring resonators for biochemical sensing applications,” IEEE J. Sel. Top. Quantum Electron. 12(1), 134–142 (2006). [CrossRef]
43. C. A. Barrios, K. B. Gylfason, B. Sánchez, A. Griol, H. Sohlström, M. Holgado, and R. Casquel, “Slot-waveguide biochemical sensor,” Opt. Lett. 32(21), 3080–3082 (2007). [CrossRef]
44. N. Chen, B. Yun, and Y. Cui, “Cladding mode resonances of etch-eroded fiber Bragg grating for ambient refractive index sensing,” Appl. Phys. Lett. 88(13), 133902 (2006). [CrossRef]
45. M. Abutoama and I. Abdulhalim, “Self-referenced biosensor based on thin dielectric grating combined with thin metal film,” Opt. Express 23(22), 28667–28682 (2015). [CrossRef]
46. B. W. Wang, Y. Zhou, Z. H. Guo, and X. Wu, “Design for Distributed Feedback Laser Biosensors Based on the Active Grating Model,” Sensors 19(11), 2569 (2019). [CrossRef]
47. S. M. Norton, T. Erdogan, and G. M. Morris, “Coupled-mode theory of resonant-grating filters,” J. Opt. Soc. Am. A 14(3), 629–639 (1997). [CrossRef]