Open Access
Add to BibSonomy
Abstract
We report on silicon waveguide distributed Bragg reflector (DBR) cavities hybridized with a tellurium dioxide (TeO2) cladding and coated in plasma functionalized poly (methyl methacrylate) (PMMA) for label free biological sensors. We describe the device structure and fabrication steps, including reactive sputtering of TeO2 and spin coating and plasma functionalization of PMMA on foundry processed Si chips, as well as the characterization of two DBR designs via thermal, water, and bovine serum albumin (BSA) protein sensing. Plasma treatment on the PMMA films was shown to decrease the water droplet contact angle from ∼70 to ∼35°, increasing hydrophilicity for liquid sensing, while adding functional groups on the surface of the sensors intended to assist with immobilization of BSA molecules. Thermal, water and protein sensing were demonstrated on two DBR designs, including waveguide-connected sidewall (SW) and waveguide-adjacent multi-piece (MP) gratings. Limits of detection of 60 and 300 × 10−4 RIU were measured via water sensing, and thermal sensitivities of 0.11 and 0.13 nm/°C were measured from 25–50 °C for SW and MP DBR cavities, respectively. Plasma treatment was shown to enable protein immobilization and sensing of BSA molecules at a concentration of 2 µg/mL diluted in phosphate buffered saline, demonstrating a ∼1.6 nm resonance shift and subsequent full recovery to baseline after stripping the proteins with sodium dodecyl sulfate for a MP DBR device. These results are a promising step towards active and laser-based sensors using rare-earth-doped TeO2 in silicon photonic circuits, which can be subsequently coated in PMMA and functionalized via plasma treatment for label free biological sensing.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
Corrections
5 April 2023: A minor correction was made to the author list.
1. Introduction
Portable diagnosis for antigens including viruses, DNA, proteins, and a variety of other biologically relevant species has become a necessity to limit the spread of disease and avoid future pandemics. To achieve this, a cost-effective solution which is mass-producible and allows for diagnostics in the field is required instead of relying on expensive laboratory facilities. Integrated optical waveguide technology enables this, by providing compact lab-on-chip biological and environmental sensors, capable of sensitive, real-time, and rapid assessment.
Amongst the photonic material platforms, silicon-on-insulator (SOI) has proven itself as a cornerstone of the field and reliable choice for waveguide sensor devices [1–4]. Repeatable and low-cost SOI photonic devices are possible by leveraging complementary metal-oxide-semiconductor (CMOS) fabrication processes and have demonstrated a variety of sensing capabilities [5]. Integrated photonic sensors are now considered competitive alternatives to laboratory and healthcare optical sensor standards such as surface plasmon resonance techniques [6]. In integrated photonic biosensors, refractive index, physical, and thermal fluctuations can all be monitored by changes in the transmission of an optical resonator, which when interrogated by a laser in the presence of an analyte can provide real-time rapid monitoring and sensing for a variety of biologically relevant species and processes. These devices have been realized on the Si photonic platform as a variety of resonators including rings, disks and interferometers [3], as well as 1D photonic crystal rings [7], and Bragg grating waveguides [5]. While rings and disks are highly compact, they suffer from drawbacks such as low free-spectral range and thus limited dynamic sensing capability, and reduced sensitivity due to limited surface area coverage. On the other hand, distributed Bragg gratings provide stopbands with dynamic and unique non-repeating resonances and are scalable in terms of sensing area, which has enabled their application as high quality evanescent waveguide sensors [8]. However, to date primarily passive waveguide and cavity structures have been explored for sensing on the SOI platform and extensive optimization has led to them approaching their fundamental limits in terms of performance metrics [5].
Laser-based sensing is an alternative approach which relies on changes in the ultra-narrow emission line(s) of an active resonator instead of shifts in the transmission of a passive structure. It can potentially allow for the detection of smaller single particles, free space or far-field interrogation, and intensity-based sensing [9], which in comparison to passive refractive index sensing devices can offer enhanced sensitivity and measurement capabilities. Rare earth ion doped glass materials have been applied in monolithic integrated lasers, are low cost, and offer wavelength versatility, making them an attractive option for silicon-based active sensing platforms [10–12]. Various demonstrations of laser based sensors have emerged, including Er3+ and Yb3+ doped SiO2 microtoroids [13,14] and patterned rare-earth doped Al2O3 to form DFB cavities [15] and ring resonators [16] for biological and environmental monitoring. Implementing these materials on an SOI chip, where the laser resonator is integrated directly with the silicon waveguide layer and the photonic sensor is not a stand-alone device but co-integrated with other system elements such as detectors, can allow for an ultra-compact form-factor and optical sensing and electronic readout all on the same chip. Significant efforts on the Al2O3 platform, which is a well-established rare earth host material, have led to demonstrations of passive and active Yb3+ doped sensors [17] with limits of detection of ∼1.0 and 3.7 × 10−6 respectively, as well as self-referenced rings with PMMA integrated gratings [18]. TeO2 is another attractive rare-earth host material for silicon photonics due to its relatively higher refractive index (n ∼2.0–2.1), low loss, relatively high rare earth solubility, and low temperature deposition [19,20]. TeO2 has been applied in on-chip optical amplification and lasing directly on silicon [11], which makes it prospective for laser-based SOI sensors. In addition to its recent demonstration as a cladding for high quality factor resonators [21] and monolithic laser material on SOI via incorporation of rare-earth dopants [11], the high refractive index of the TeO2 cladding enables the mode to be ‘pulled’ up to the top layer where sensing occurs for better sensitivity. A TeO2 microcavity coupled to a Si waveguide has been demonstrated as a sensor platform, which however suffered from bus coupling control and limited sensing surface area [22]. Si-TeO2 hybridized DBR waveguides might provide a promising alternative pathway as a biological sensor and have promise in developing as an on-chip laser with scalable area and dynamic sensitivity for sensing.
In both passive and laser-based sensors, surface layer functionalization is important for facilitating biological interactions with proteins or viruses including immobilization, binding, and/or adsorption (physisorption and chemisorption). Additional surface chemistry is required for these interactions, which can be achieved through functionalization via the addition of oxygen rich groups which influence bonding sites for the biomarkers. Rather than performing functionalization directly on the Si or oxide cladding surface, which may lead to deleterious optical effects such as loss in the resonator, materials like polymers may be coated on top of the waveguides and functionalized by a variety of chemical or plasma methods. PMMA is a prospective material for integration into sensor designs due to its high transparency in the near infrared, relatively low cost, and proven capability for functionalization via oxygen plasma treatment [23]. Plasma treatment in this context acts as a straightforward way to allow O2 to react with the surface of the PMMA, which enables sites for promoting physical adsorption of biomarkers such as proteins. PMMA has been applied in integrated optical sensors, including as a fluorescently tagged biological sensing material for the detection of Chlamydia trachomatis specific immunoglobulins [24], but to our knowledge oxygen plasma PMMA functionalization has not been demonstrated directly on SOI biosensors.
Here we report on the design, fabrication, optical properties, and sensor results of functionalized PMMA coated hybrid Si-TeO2 waveguides with distributed Bragg reflector (DBR) cavities. Two Si DBR designs are characterized and compared with a variety of cladding materials including PMMA, TeO2 and CYTOP, as well as a combination of deposited TeO2 and spin-coated PMMA. A fabrication process for directly functionalizing Si-TeO2/PMMA sensors for chemisorption is demonstrated as well as subsequent thermal, water, and protein sensing using BSA. It is demonstrated that plasma functionalization enables direct immobilization and sensing of BSA on the PMMA surface and is not deleterious to the optical performance of the Si-TeO2 DBR resonator. These results demonstrate a promising platform for passive and rare-earth laser-based sensing on silicon photonic chips and pathways for optimizing the design moving forward.
2 Design and fabrication
2.1 Reactive magnetron sputtering of TeO2 on Si foundry DBR resonators
The uncoated sensor chips were fabricated using the Advanced Micro Foundry (AMF) silicon photonics fabrication process, with 0.22 µm thick × 0.5 µm wide silicon strip waveguides on 2.0 µm of buried oxide on silicon substrates. Two DBR resonator designs were selected to study the sensor performance when adjusting properties such as the optical mode overlap with the sensing medium and grating structure: one with sidewall (SW) corrugations of 0.6 µm and a period of 0.326 µm in a 0.5 µm wide Si waveguide, and one with multiple 0.14 µm wide grating pieces with a period of 0.341 µm and a 0.25 µm gap adjacent to a 0.4 µm wide Si waveguide. The periods were selected to obtain resonances for the transverse electric (TE) polarized fundamental mode around 1550 nm. Gradual transitions of 50 µm length were included for the multi-piece (MP) bus width taper from 0.5 to 0.4 µm and bus-grating gap which reduces from 0.8 µm to 0.25 µm. Cavity lengths of 1200 and 1100 µm and symmetric gratings lengths of 500 and 600 µm were used in the SW and MP design, respectively. To improve edge coupling to the waveguide, inverse tapering was used to expand the mode which consists of 50 µm long transitions to a 0.18 µm waveguide width at the facet. SEM images of each DBR type before depositing TeO2 with inset cross section drawings and images of the transition sections are shown in Fig. 1(a) and (b), and schematics and details of the design of the resonators are displayed in Fig. 1(c) and (d), respectively. SW DBR waveguide resonators were selected for investigation based on their well understood operation for single mode passive resonator devices and sensors [5]. It was anticipated that the MP design would demonstrate resonances with decreased bandwidth due to lighter perturbations to the waveguide mode [25], which is advantageous for the realization of sensors and laser cavities due to sharper, more selective resonance spectra. The two designs represent distinct variations of the DBR strip waveguide and allow for a comparison between cavities with high grating strengths and Si overlap (SW design) vs. cavities with larger modes, weaker grating perturbations and higher overlap in the surrounding materials (MP design).
Fig. 1. SEM images of (a) sidewall (SW) and (b) multi-piece (MP) silicon DBR waveguides with insets showing transition sections and cross section diagrams of the waveguide grating structures. (c) Design schematic and (d) specifications of SW and MP DBR gratings.
We deposited 0.5 ± 0.02 µm of TeO2 as an end of line process using reactive radio frequency (RF) magnetron sputtering after receiving the uncladded SOI chips from the foundry following the procedure outlined in [26]. The process was carried out at ambient temperature and a chamber pressure of 3 mTorr in oxygen and argon ambient, with a deposition rate of ∼23 nm/min. The experimental procedures as described in [26], including prism coupling loss measurements and fine tuning of the deposition recipe, were used to optimize the deposited films for low optical propagation loss. In past works using the same process, SEM images are included which demonstrate the TeO2 thickness and conformality to an etched oxide trench [22] and silicon nitride waveguides [26,27]. By measuring the thickness of deposited layers on Si witness samples which were included in the chamber during deposition via variable angle spectral ellipsometry (VASE), the thickness on fabricated sensors can be estimated post deposition. To better control the thickness of the deposited TeO2 layer, in-situ optical thickness monitoring is suggested during deposition for large-scale production. Low temperature deposition of the TeO2 film allows for integration with thermally sensitive materials such as metals which are applied in electrical contacts for thermal tuners and photodetectors in SOI-based sensor circuits.
2.2 Polymer spin coating and functionalization via oxygen plasma
PMMA spin coating was carried out at 1750rpm for 45 seconds with a bake temperature of 80 °C for 1.5 minutes followed by a cure step of 150 °C for 10 minutes on TeO2 coated chips. To calibrate the process, VASE was used to measure the thickness on Si samples by testing different spin speeds while coating. This resulted in samples with coating thicknesses that had a repeatable margin of ± 0.01 µm. To functionalize the PMMA layer and activate O2 groups on the surface, oxygen plasma treatment was carried out with a Harrick plasma cleaner at 750 mTorr for 1 min with 27.5 W of RF power. This process was repeated for any chips which required subsequent measurements after submersion in liquid and air drying or significant time spent out in open air. The stability of the process is time sensitive and known to decay once exposed to ambient conditions for approximately a week [23]. Immediate PDMS or SU8 microfluidic capping may protect the functional groups from interacting with the environment and decreasing in surface energy, as SU8 has shown to provide more stability in surface energy in ambient conditions after functionalization [23]. In this manner, the sensing layer could be protected from the environment or analyte until a controlled delivery method is carried out, allowing for increased stability of the packaged sensor.
To quantify the impact of plasma exposure for functionalization, liquid drop water contact angle measurements were carried out on PMMA films on 3-inch Si wafers using a ramé-hart goniometer characterization system following the procedure outlined in [28]. The plasma exposure was carried out using O2 at a pressure of 750 mTorr with 27.5 W of forward power at varying times. The O2 plasma treatment was anticipated to impart oxygen-based functional groups with increased energy on the PMMA surface, which in turn can increase hydrophilicity. Increased hydrophilicity can be observed by a decreased contact angle when a water droplet comes into contract with the top surface. The results are shown in Fig. 2, demonstrating a ∼40° decrease in contact angle after 5 minutes of plasma treatment which shows a distinct transition from a hydrophobic to a hydrophilic surface. This increase in surface hydrophilicity confirms the modification of the PMMA by plasma treatment and likely plays a role in the protein binding mechanisms which can vary for hydrophilic and hydrophobic surfaces, along with the introduction of functional groups at the surface [29]. Additionally, VASE was used to measure the thickness of the PMMA layer before and after plasma treatment to ensure etching was not occurring during the process. No change in thickness was observed within the error limit of the VASE measurements.
Fig. 2. Water droplet contact angle as a function of oxygen plasma treatment for 0.2-µm-thick PMMA films on silicon substrates with inset drawings illustrating different contact angles.
Figure 3 summarizes the overall fabrication process of the DBR waveguide sensor. To investigate the DBR response for different cladding refractive indices, DBR devices were also coated with PMMA and CYTOP without any TeO2 film. CYTOP spin coating was carried out at 1750rpm for 45 seconds, with a bake and cure time of 120 and 180 °C for 1.5 and 10 minutes respectively. An isometric design view of the final proposed device with labelled materials, overlaid optical mode, and illustrated analyte particles is shown in Fig. 4. The propagation direction for the mode is shown, with oscillations extending into the top and bottom layers to illustrate the shared overlap with immobilized proteins on the surface.
Fig. 4. An isometric view of a Si/TeO2 DBR waveguide sensor coated in PMMA with labelled materials and overlapped fundamental TE mode with illustrated attached molecules and optical propagation direction and Bragg shifted frequency indicated by coloured arrows for integrated biological sensing.
3 Characterization
3.1 DBR optical transmission measurements and calculations
The edge coupling setup shown in Fig. 5 was used to characterize the DBR cavities. Manual micrometer-controlled stages were used to launch 1510–1640 nm laser light from 2.5 µm spot size lensed fibers. Initially, uncladded chips received from foundry were measured to investigate the impact of fabrication-related variation in Si waveguide features from the multi-project wafer (MPW) run. Also, measurements before and after post-processing can be used to track the resonances and test the devices sensitivity to cladding. The measured resonances for air cladded SW gratings are shown in inset of Fig. 6, demonstrating the variation among identically designed devices on different chips from the same SOI wafer. Resonances were observed in two MP samples demonstrating an average central wavelength of 1510.8 nm and extinction ration (ER) of 11.8 as shown in Fig. 6, but others are anticipated to be resonating at a lower wavelength than 1510 nm outside the measurement range of the tunable laser used.
Fig. 5. Illustration of the fiber coupling setup used for DBR transmission measurements and image of the sample during test.
Fig. 6. Comparison of spectra collected from SW and MP Si DBR cavities with different top-cladding materials, including air, CYTOP, PMMA and TeO2 with thicknesses of N/A, 0.5, 0.8 and 0.9 µm, respectively.
After measuring resonances with air cladding, selected samples were either spin coated with 0.8 µm of PMMA, 0.95 µm of CYTOP or sputter coated with 0.5 µm of TeO2 to investigate the SW and MP DBR transmission spectra with different top-cladding refractive indices. These measurements are also shown in Fig. 6, with inset zoomed-in views of the MP intra-band DBR resonances. Figure 7 demonstrates the transmission from fully fabricated samples with 0.5 µm of deposited TeO2 and 0.2 µm of subsequently spin-coated PMMA. The calculated quality factors are indicated for the samples intended for subsequent sensor measurements. The quality factors in Fig. 7 are calculated using the ratio of the resonance wavelength to the full width at half maximum (FWHM) for intra-band resonances, which have bandwidths in the pm range. We observe sharp and narrow resonance spectra for MP gratings with wider and deeper resonances for SW DBRs. In some cases, split resonances are observed as is the case for CYTOP in Fig. 6. This can be due to counter propagating modes of slightly varying loss, asymmetry in the fabrication process or a variety of other effects from processing the samples. The measured central wavelength varies within the S, C and L bands around the designed central wavelength of 1550 nm.
Fig. 7. Transmission spectra for TeO2- and PMMA-coated SW and MP Si DBR cavities on two samples, used for water sensing and protein sensing separately. The insets show close-up views of the resonances. The quality factors are calculated using the ratio of the central wavelength to the FWHM of the indicated resonances with black dotted circles.
The TeO2- and PMMA-coated DBR cavity devices corresponding to the measurements in Fig. 7 were treated with oxygen plasma to investigate its influence on their transmission properties. A 1-minute exposure was used for the DBR chip treatment based on the limited change in contact angle for longer times and because additional plasma exposure times might damage the TeO2 layer beneath the relatively thin PMMA layer. The samples were measured before and immediately after functionalization and the optical transmission spectra of the DBRs were found to be unaffected by the plasma treatment. By fitting the resonances of the DBR transmission spectra shown in Fig. 7 before and after treatment and obtaining the quality factors, which are directly connected to the waveguide losses, it was confirmed that significant surface scattering or absorption losses were not introduced. These results suggest the imparted chemical functionalization of the surface layer of the PMMA is not deleterious to the optical performance of our sensors.
The calculated optical properties of the fabricated DBR cavities, including the mode profiles, effective indices, optical intensity overlap with the different waveguide materials, and effective modal areas are plotted in Figs. 8(a) and (b), (c) and (d), (e) and (f), and 8(c) and (d) (insets), respectively. The cross-sectional mode profiles were calculated assuming conformal TeO2 and PMMA coating of the Si waveguide only (and not the grating pieces) for a straightforward and close approximation of the mode properties. In the plots the TeO2 film thickness is varied in order to consider potential active sensor designs, which would require careful optimization of the mode overlap with both the gain medium (rare-earth doped TeO2) and the sensing medium (PBS). The higher effective index in the SW DBR designs is due to the mode being more confined to the Si waveguide, which also leads to a smaller modal area and less overlap with the top PMMA layer and analyte for sensing. In contrast, the MP grating pieces pull the mode further away from the bus causing a larger modal area and subsequent higher modal overlap in the top sensing layer. A difference in overlap can also be observed between the Si gratings and TeO2 layer, which vary from 3.0 and 19.5% to 0.53 and 30.4% partial power overlap between the SW to MP design, respectively, when coated with 0.5 µm of TeO2. Although a longer cavity is needed to achieve comparable grating strengths to the SW design, the MP design provides a useful tradeoff for active sensors where not only the surface overlap is crucial for performance, but the hybrid gain material surrounding the Si waveguide core is required to have high optical overlap [11]. By tuning parameters like the Si bus width, DBR grating piece width and gap, TeO2 thickness, and PMMA thickness the design can be optimized for increased sensitivity and overlap with the analyte or in this case PBS liquid layer.
Fig. 8. Transverse electric field profile for (a) SW and (b) MP grating designs for a TeO2 and PMMA thickness of 0.5 and 0.2 µm, respectively. Effective indices for (c) SW and (d) MP waveguide gratings for the first four modes with inset fundamental modal area for varying TeO2 thickness. Partial power overlap for the fundamental transverse electric (TE0) mode in (e) SW and (f) MP waveguide grating materials and inset closeup of PBS sensing fluid, PMMA, and grating overlaps for TeO2 film thicknesses of interest.
The grating strengths were obtained following the method outlined in [30] based on the penetration length of the mode into the grating for Al2O3 DBRs. Transfer matrix methods (TMM) were used to simulate the reflectivity spectrum of the resonators, and to obtain the ERs and central wavelengths of the devices. Deviations from the nominal design parameters such as reducing the grating feature size were required to match the measured data, which suggests that rounding of the features occurred during lithography. Additionally, as mentioned, a conformal fill and coating has been approximated in the grating gaps, as well as homogenous thickness for the various layers applied with spin-coating. For SW devices, the calculated ERs were much higher than the measured values (+50 dB) suggesting resonances were not fully resolved during measurement. The simulated responses for MP DBRs match well with measured data, with slight deviations which can be attributed to uncertainties in the layer thicknesses, and the assumption of conformal fill between the multi-piece grating features and the Si bus waveguide during simulation. Figure 9 shows an example of the simulated response overlapped with a measurement for the TeO2/PMMA coated MP resonator used for protein sensing shown in Fig. 7. Table 1 summarizes the calculated and measured grating properties for each structure shown in Figs. 6 and 7.
Table 1. Summary of calculated and measured properties of fabricated SW and MP DBR waveguides.
3.2 Thermal sensing
To investigate the Si-TeO2/PMMA DBR resonator as a temperature sensor, a sample was placed on a stage with a copper mount and Peltier cooler while the resonance was tracked during heating from 25 to 50 °C. To obtain each data point, a feedback circuit maintained the desired stage temperature for 5 minutes before taking an optical measurement. The results are displayed in Fig. 10, with a photograph of the sample during test shown in the inset.
Fig. 10. Central resonant wavelength shift as a function of temperature for SW and MP waveguide gratings and inset photograph of sample during test with TEC element and copper stage for thermal control.
Figure 10 demonstrates a thermal sensitivity which is almost identical for both designs, but higher for the SW case, where the optical overlap with the Si layer and grating pieces is higher, and therefore has more interaction with the high thermo-optic coefficient Si material. The thermal resonance shift is caused by a combination of the materials’ thermal expansion and the overall thermo-optical effect which is primarily influenced by the thermo-optic coefficients and mode overlap with the Si and TeO2 layers. The thermo-optic coefficients of bulk Si and TeO2 are 1.8 and 0.7 × 10−4 respectively. The other materials such as SiO2 and PMMA do not have a significant impact because of their relatively lower thermo-optic coefficients. The thermo-optic coefficient (σT) of the sensors was calculated using the measured thermal sensitivity (ST = Δλ/ΔT), measured resonant wavelength (λ0), and calculated group index (ng) which depends on the calculated dispersion of the fundamental mode: ng = neff + λ0(dneff/dλ). This yields a result of ng = 3.39 and 3.22, and subsequently using σT = ST(ng/λ0), leads to a σT of 2.74 and 2.23 × 10−4 °C−1 at 1610 and 1590 nm for SW and MP gratings, respectively. For both devices this is higher than the thermo-optic coefficient of Si, and well above results reported for TeO2 [31], and therefore might include additional contributions from material and structural deformation under heating. This could be caused by nonconformal coating of the layers on the grating sections leading to voids in the film stack. These voids would allow for additional degrees of freedom for thermal expansion and higher thermal sensitivity than the bulk material. Although PMMA and TeO2 have less prominent contributions to thermal sensitivity on this platform, their thermal stability is the limiting factor in applications of these sensors in high temperature environments. TeO2 and PMMA have known temperature limits of approximately 200°C [32,33], which still makes them suitable for applications in the biological domain where temperatures at these scales are not relevant for the study of biological interactions. For this reason, the temperatures used during measurement were kept between 25–50°C, as the sensors are not intended for environments with broad temperature ranges. Here, a relatively high thermal sensitivity is demonstrated compared to other TeO2 integrated sensors [22], and Bragg gratings [34,35], on the SOI platform, which can be considered an advantage or drawback depending on the sensing application and the relative importance of noise factors and environmental fluctuations. In future designs with higher optical overlap in the TeO2 and sensing layer it is anticipated that lower thermal sensitivities can be achieved.
3.3 Liquid sensing
Liquid sensing was carried out with deionized (DI) water and PBS as a solution for diluting BSA protein. As shown in many demonstrations [1,2,4], PDMS or PMMA cap layers or SU-8 microfluidic structures can selectively deliver liquid to channels and specific waveguides or arrays on the chip. These delivery channels can in theory be easily integrated into our sensing scheme with PMMA already being the top layer and acting as a cap layer for a functionalized surface. However, for the purposes of this study, liquids were simply dropped onto the surface of the sample during test without control of the flow or volume of liquid. DI water was used to determine the sensors’ bulk sensitivity using the well-known relation Sbulk = Δλ/Δnliquid, where Δnliquid refers to the change in index between water and air (1.316-1.0). Transmission measurements before and after DI water application yielded sensitivities of 10.2 and 0.16 nm/RIU from a shift of 3.3 and 0.05 nm for MP and SW DBRs, respectively. The reduced water droplet contact angle from plasma treatment did not lead to an increased shift from evanescent interaction with the DBR mode compared to no plasma treatment, because the droplet covered the entire device in both cases. However, the plasma enabled hydrophilicity was observed during measurement, as the water did not bead to the surface of the PMMA, but rather formed a film on the top of the sensor after treatment. The water sensing measurement process is depicted in Fig. 11(a), while Fig. 11(b) shows the transmission measurements before and after plasma treatment with water coverage, and Fig. 11 c) shows images of the sample in air and during water sensing.
Fig. 11. (a) Cross section drawings showing different surface treatments for DI water sensing and (b) corresponding labelled transmission spectra obtained from the MP DBR device, showing a shift of 3.3 nm in water with and without plasma treatment. (c) Images of the chip with plasma treatment during measurement in air and water.
The MP DBRs demonstrated two orders of magnitude higher sensitivity compared to SW DBRs under water exposure. This can be attributed to the larger modal overlap with the water (0.064 vs. 0.016% calculated for MP and SW, respectively). When considering future sensor devices with similar designs and rare-earth doped oxide layers that also operate as a laser, excess loss due to increased overlap with the analyte, especially water, needs to be considered and balanced in the design in order to maintain a high quality factor (Q). To determine the limit of detection (LoD), the relation LoD = 1/QSbulk was used. The MP and SW DBRs demonstrated limits of detection of 60.0 and ∼300 × 10−4 RIU, respectively. Table 2 compares the figures of merit and properties of various Si DBR and distributed feedback (DFB) devices used in liquid sensing from literature, including their waveguide structure and material, wavelength and polarization, sensing analyte, sensing quality factor, sensitivity, and LoD. Here the sensitivity is compared based on bulk sensing measurements, such as immersion in water, where the biological analyte is also listed to demonstrate the device was used for surface sensing. The sensitivities and limits of detection reported here are lower than state-of-the-art silicon photonic evanescent field sensors shown in the table. However, we show similar or higher Q factors which is important for active sensors and the addition of TeO2 and PMMA layers demonstrates a functional hybrid sensor platform which can in future allow for rare-earth laser biological sensors on the SOI platform.
Table 2. Various Si DBR and DFB sensing demonstrations (*calculated).
3.4 Protein sensing
Protein sensing was investigated using BSA in PBS buffer solution. A PBS baseline was first measured before incubation with the BSA, which showed a permeant change in the resonance (∼1.0 nm redshift) of the MP DBR Si-TeO2/PMMA sensor. 2.0 µg/mL of BSA was then diluted into the PBS buffer solution and a sample was incubated in it for 1 hour, without any prior oxygen plasma functionalization. The chip was then measured after being flushed with PBS to ensure excess proteins were removed and only proteins attached to the surface were measured. The chip was also measured while submerged in PBS, to isolate the impact of the proteins on the optical spectrum. This resulted in a negligible shift of ∼0.1 nm within the limits of thermal fluctuations. After cleaning with isopropanol and drying with nitrogen, the incubation and measurement process was then repeated after performing oxygen plasma treatment on the sample. This resulted in a 1.6 nm shift attributable to protein immobilization on the surface of the functionalized PMMA for the MP DBR. Sodium dodecyl sulfate (SDS) was used afterwards to strip the proteins from the surface of the sensor and the chip was cleaned with isopropanol and dried with nitrogen, which demonstrated a recovery of the initial baseline resonant wavelength. Two MP samples were tested, and the measurements confirmed repeatability of the result and the requirement of O2 plasma functionalization for BSA immobilization, as well as the requirement for SDS to strip the proteins, which normal cleaning methods such as isopropanol would not reproduce. The secondary sample used was designed identically and resulted in a shift of ∼1.4 nm. The discrepancy between samples is likely due to varying deposited film heights from separate depositions and PMMA spun thicknesses, and thus varying sensitivities between the two samples. No noticeable shift was detected for the SW resonances. This is due to the lower sensitivity in SW devices as shown via water sensing. The protein sensing measurement process and the results from each step are displayed for the MP DBR sensor which displayed a 1.6 nm shift in Fig. 12.
Fig. 12. Central resonant wavelength shift for different measurement steps during protein sensing, including top layer changes between 1) air, 2) PBS baseline, 3) PBS after PBS + BSA immersion, 4) air after acetone clean and O2 plasma functionalization, 5) PBS, 6) PBS after PBS + BSA immersion with functionalization and protein immobilization, 7) air after SDS immersion and protein stripping, and 8) air repeated 1 hour after.
The plasma functionalized PMMA layer used here can remove the need for an initial chemical functionalization step, which sometimes requires biotin for selective attachment or harmful chemicals such as toluene, and is distinct from other functionalization methods applied in biological sensing studies on Si [36]. This manner of functionalization allows for the direct chemisorption and immobilization of blocking proteins by the interaction of molecules such as BSA to the oxygen-based functional sites provided by the plasma treated PMMA surface. Compared to physisorption, this functionalization strategy and immobilization through chemisorption allows for a decreased chance for the proteins to denature on the surface [24]. This provides a more robust and stable surface for the subsequent detection of antibodies, viruses and other biologically relevant species which require detection in biological fluids such as blood [40]. With the added ability of hydrophilicity control, O2-plasma functionalized PMMA’s inclusion in the hybrid sensor stack enables liquid-based sensing for the biological domain using resonators with active sensing potential. Direct plasma functionalization of PMMA as a top layer on Si-TeO2 DBR resonators provides a novel platform for the sensing of proteins via specific and non-specific binding in silicon photonic circuits.
Future work will focus on measurement technique optimization in order to properly assess changing concentrations and monitoring of resonant wavelengths in real time. Real-time measurements, control of the flow and concentration and monitoring of the binding events for the BSA will allow for a concentration minimum study [36] or Langmuir dissociation constant calculation [41]. In addition, improved hybrid sensor designs and active resonators will be investigated based on this initial demonstration. Using plasma functionalized PMMA as a biological interface layer in combination with rare-earth doped TeO2 is a novel prospective approach to hybrid laser sensors for silicon photonic platforms. More generally, a direct approach of functionalization to achieve immobilization and detection of proteins on a polymer surface opens many pathways for novel handheld silicon photonic devices and sensors.
4. Conclusion
We have demonstrated the fabrication and characterization of a PMMA coated Si-TeO2 hybrid DBR waveguide sensor for biological applications. Two Si DBR designs were designed and fabricated using a standard silicon photonic foundry process which enables low cost, volume production. The transmission response of the DBRs was investigated around 1550 nm for a variety of top-cladding materials and the quality factors for PMMA coated Si-TeO2 devices were measured yielding values of 1.82 and 1.12 × 105 for SW and MP devices, respectively. Plasma functionalization of the PMMA surface was carried out and characterized via water droplet contact angle and transmission measurements, demonstrating a decrease in angle from 70 to 35° and minimal effect on the optical spectra of the DBRs, respectively. Thermal sensing was performed demonstrating a shift of 0.13 and 0.11 nm/°C for SW and MP DBRs respectively, over a temperature range of 25–50 °C. Liquid sensing was performed with DI water in order to determine the bulk sensitivities of the MP and SW devices. Resonance shifts of 0.05 and 3.2 nm were measured, which correspond to sensitivities of 0.16 and 10.2 nm/RIU for SW and MP DBRs respectively. MP DBR devices were then used to demonstrate the immobilization of blocking proteins using PBS as a buffer solution with a BSA concentration of 2 µg/mL. It was observed that plasma functionalization is required to detect a shift after incubation with the protein solution, suggesting the binding is taking place on the PMMA functionalized surface via chemisorption. Using an SDS stripping solution, the sensor’s original resonant condition was recovered after sensing, suggesting potential re-use of single devices. These Si-TeO2 hybridized DBRs coated in plasma treated PMMA demonstrate a pathway towards improved functionalized devices for protein sensing applications and the potential for active resonator sensors based on rare-earth-doped TeO2 layers on silicon photonic chips.
Funding
Satellite Canada Innovation Network (HTSN 611, HTSN-621); Ontario Ministry of Research and Innovation (ER15-11-123); Natural Sciences and Engineering Research Council of Canada (RGPIN-2019-05994); Canada Foundation for Innovation (35548); Natural Sciences and Engineering Research Council of Canada (RGPIN-2017-06423, RGPIN-2022-05258).
Acknowledgments
We thank CMC Microsystems and the SiEPIC Program for facilitating the silicon photonics fabrication, the Centre for Emerging Device Technologies (CEDT) at McMaster University for support with the cleanroom and reactive sputtering system, and The Canadian Centre for Electron Microscopy (CCEM) for support with scanning electron microscopy.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data available on request from the authors.
References
1. L. Chrostowski, S. Grist, J. Flueckiger, W. Shi, X. Wang, E. Ouellet, H. Yun, M. Webb, B. Nie, Z. Liang, K. C. Cheung, S. A. Schmidt, D. M. Ratner, and N. A. F. Jaeger, “Silicon photonic resonator sensors and devices,,” Laser Reson. Microresonators, Beam Control XIV 8236, 823620 (2012). [CrossRef]
2. E. Luan, H. Shoman, D. M. Ratner, K.C. Cheung, and L. Chrostowski, “Silicon photonic biosensors using label-free detection,” Sensors 18(10), 1–42 (2018). [CrossRef]
3. K. De Vos, I. Bartolozzi, E. Schacht, P. Bienstman, and R. Baets, “Silicon-on-Insulator microring resonator for sensitive and label-free biosensing,” Opt. Express 15(12), 7610–7615 (2007). [CrossRef]
4. A. F. Gavela, D. G. García, J. C. Ramirez, and L. M. Lechuga, “Last advances in silicon-based optical biosensors,” Sensors 16(3), 1–15 (2016). [CrossRef]
5. S. Schmidt, J. Flueckiger, W. Wu, S. M. Grist, S. Talebi Fard, V. Donzella, P. Khumwan, E. R. Thompson, Q. Wang, P. Kulik, X. Wang, A. Sherwali, J. Kirk, K. C. Cheung, L. Chrostowski, and D. Ratner, “Improving the performance of silicon photonic rings, disks, and Bragg gratings for use in label-free biosensing,” Biosensing Nanomedicine 7, 916660M (2014). [CrossRef]
6. P. Steglich, G. Lecci, and A. Mai, “Integrated circuit (PIC) biosensors: a comparative review,” Sensors 22(8), 1–19 (2022). [CrossRef]
7. S. M. Lo, S. Hu, G. Gaur, Y. Kostoulas, S. M. Weiss, and P. M. Fauchet, “Photonic crystal microring resonator for label- free biosensing,” Opt. Express 25(6), 7046–7054 (2017). [CrossRef]
8. S. Kaushal, R. Cheng, M. Ma, A. Mistry, M. Burla, L. Chrostowski, and J. Azaña, “Optical signal processing based on silicon photonics waveguide Bragg gratings: review,” Front. Opt. 11(2), 163–188 (2018). [CrossRef]
9. T. Reynolds, N. Riesen, A. Meldrum, X. Fan, J.M.M. Hall, T. M. Monro, and A. François, “Fluorescent and lasing whispering gallery mode microresonators for sensing applications,” Laser Photonics Rev. 11(2), 1–20 (2017). [CrossRef]
10. L. Agazzi, J. D. B. Bradley, M. Dijkstra, F. Ay, G. Roelkens, R. Baets, K. Wörhoff, and M. Pollnau, “Monolithic integration of erbium-doped amplifiers with silicon-on-insulator waveguides,” Opt. Express 18(26), 27703 (2010). [CrossRef]
11. K. M. Kiani, H. C. Frankis, C. M. Naraine, D. B. Bonneville, A. P. Knights, and J. D. B. Bradley, “Lasing in a hybrid rare-earth silicon microdisk,” Laser Photonics Rev. 16(1), 2100348 (2022). [CrossRef]
12. N. Li, M. Xin, Z. Su, E.S. Magden, N. Singh, J. Notaros, E. Timurdogan, P. Purnawirman, J. D. B. Bradley, and M. R. Watts, “A silicon photonic data link with a monolithic erbium-doped laser,” Sci. Rep. 10(1), 1–9 (2020). [CrossRef]
13. L. He, Ş. K. Özdemir, J. Zhu, W. Kim, and L. Yang, “Detecting single viruses and nanoparticles using whispering gallery microlasers,” Nat. Nanotechnol. 6(7), 428–432 (2011). [CrossRef]
14. E. P. Ostby and K. J. Vahala, “Yb-doped glass microcavity laser operation in water,” Opt. Lett. 34(8), 1153 (2009). [CrossRef]
15. E.H. Bernhardi, K.O. van der Werf, A.J.F. Hollink, K. Wörhoff, R.M. de Ridder, V. Subramaniam, and M. Pollnau, “Intra-laser-cavity microparticle sensing with a dual-wavelength distributed-feedback laser,” Laser Photonics Rev. 7(4), 589–598 (2013). [CrossRef]
16. M. de Goede, M. Dijkstra, R. Obregón, J. Ramón-Azcón, E. Martínez, L. Padilla, F. Mitjans, and S.M. Garcia-Blanco, “Al2O3 microring resonators for the detection of a cancer biomarker in undiluted urine,” Opt. Express 27(13), 18508 (2019). [CrossRef]
17. M. Goede, L. Chang, J. Mu, M. Dijkstra, R. Obregón, E. Martínez, L. Padilla, F. Mitjans, and S. M. Garcia-Blanco, “Al2O2:Yb3+ integrated microdisk laser label-free biosensor,” Opt. Lett. 44(24), 5937–5940 (2019). [CrossRef]
18. M. Goede, M. Dijkstra, L. Chang, N. Acharyya, G. Kozyreff, R. Obregón, E. Martínez, and S. M. García-Blanco, “Mode-splitting in a microring resonator for self-referenced biosensing,” Opt. Express 29(1), 346–358 (2021). [CrossRef]
19. S. J. Madden and K. T. Vu, “Very low loss reactively ion etched tellurium dioxide planar rib waveguides for linear and non-linear optics,” Opt. Express 17(20), 369–373 (2009). [CrossRef]
20. K. Vu and S. Madden, “Tellurium dioxide erbium doped planar rib waveguide amplifiers with net gain and 2.8 dB / cm internal gain,” Opt. Express 18(18), 19192–19200 (2010). [CrossRef]
21. K. M. Kiani, D. B. Bonneville, A. P. Knights, and J. D. B. Bradley, “High-Q TeO2–Si hybrid microring resonators,” Appl. Sci 12(3), 1363 (2022). [CrossRef]
22. H. C. Frankis, D. Su, D. B. Bonneville, and J. D. B. Bradley, “A tellurium oxide microcavity resonator sensor integrated on-chip with a silicon waveguide,” Sensors 18(11), 4061 (2018). [CrossRef]
23. R. Landgraf, M. K. Kaiser, J. Posseckardt, B. Adolphi, and W. J. Fischer, “Functionalization of polymer sensor surfaces by oxygen plasma treatment,” Procedia Chem. 1(1), 1015–1018 (2009). [CrossRef]
24. S. Sathish, N. Ishizu, and A. Q. Shen, “Air plasma-enhanced covalent functionalization of poly(methyl methacrylate): high-throughput protein immobilization for miniaturized bioassays,” ACS Appl. Mater. Interfaces 11(49), 46350–46360 (2019). [CrossRef]
25. X. Wang, W. Shi, H. Yun, S. Grist, N. A. F. Jaeger, and L. Chrostowski, “Narrow-band waveguide Bragg gratings on SOI wafers with CMOS-compatible fabrication process,” Opt. Express 20(14), 3013–3015 (2012). [CrossRef]
26. H. C. Frankis, K. Miarabbas Kiani, D. B. Bonneville, C. Zhang, S. Norris, R. Mateman, A. Leinse, N. D. Bassim, A. P. Knights, and J. D. B. Bradley, “Low-loss TeO2-coated Si3N4 waveguides for application in photonic integrated circuits,” Opt. Express 27(9), 12529 (2019). [CrossRef]
27. H. C. Frankis, K. Miarabbas Kiani, D. Su, R. Mateman, A. Leinse, and J. D. B. Bradley, “High-Q tellurium-oxide-coated silicon nitride microring resonators,” Opt. Lett. 44(1), 118–121 (2019). [CrossRef]
28. T. Huhtamäki, X. Tian, J. T. Korhonen, and R. H. A. Ras, “Surface-wetting characterization using contact-angle measurements,” Nat. Protoc. 13, 1521–1538 (2018). [CrossRef]
29. C. Jan van Oss, “Hydrophobicity and hydrophilicity of biosurfaces,” Curr. Opin. Colloid Interface Sci 2(5), 503–512 (1997). [CrossRef]
30. E. H. Bernhardi, “Bragg-grating-based rare-earth-ion-doped channel waveguide lasers and their applications,” University of Twente (2012).
31. T. Honma, N. Ito, T. Komatsu, and V. Dimitrov, “Thermo-optic properties and electronic polarizability in alkali tellurite glasses,” J. Am. Ceram. Soc. 93(10), 3223–3229 (2010). [CrossRef]
32. R. A. H. El-Mallawany, Tellurite Glasses Handbook: Physical Properties and Data, 2nd Edition (CRC Press2002).
33. U. Ali, K. J. Bt. A. Karim, and N. A. Buang, “A review of the properties and applications of poly (methyl methacrylate) (PMMA),” Polym. Rev. 55(4), 678–705 (2015). [CrossRef]
34. N. N. Klimov, S. Mittal, M. Berger, and Z. Ahmed, “On - chip silicon waveguide Bragg grating photonic temperature sensor,” Opt. Lett. 40(17), 4–7 (2015). [CrossRef]
35. G. Kim, H. Lee, C. Park, S. Lee, B. Lim, H. Bae, and W. Lee, “Silicon photonic temperature sensor employing a ring resonator manufactured using a standard CMOS process,” Opt. Express 18(21), 22215–22221 (2010). [CrossRef]
36. E. Luan, H. Yun, M. Ma, D.M. Ratner, K.C. Cheung, and L. Chrostowski, “Label-free biosensing with a multi-box sub-wavelength phase-shifted Bragg grating waveguide,” Biomed. Opt. Express 10(9), 4825–4838 (2019). [CrossRef]
37. X. Wang, S. Grist, J. Flueckiger, N. A. F. Jaeger, and L. Chrostowski, “Silicon photonic slot waveguide Bragg gratings and resonators,” Opt. Express 21(16), 19029–19039 (2013). [CrossRef]
38. S. T. Fard, S. M. Grist, V. Donzella, S. A. Schmidt, J. Flueckiger, X. Wang, W. Shi, A. Millspaugh, M. Webb, D. M. Ratner, K. C. Cheung, and L. Chrostowski, “Label-free silicon photonic biosensors for use in clinical diagnostics,” Silicon Photonics VIII 8629, 862909 (2013). [CrossRef]
39. A. S. Jugessur, J. Dou, J. S. Aitchison, R. M. De La Rue, and M. Gnan, “A photonic nano-Bragg grating device integrated with microfluidic channels for bio-sensing applications,” Microelectron. Eng. 86(4-6), 1488–1490 (2009). [CrossRef]
40. K. N. Sask, “Antithrombogenic biomaterials: surface modification with an antithrombin- heparin covalent complex,” McMaster University (2018).
41. Y. L. and L. Shen, “From Langmuir kinetics to first- and second-order rate equations for adsorption,” Langmuir 24(20), 11625–11630 (2008). [CrossRef]
