We demonstrate optical fiber-pigtailed temperature sensors based on dielectric-loaded surface plasmon-polariton waveguide-ring resonators (DLSPP-WRRs), whose transmission depends on the ambient temperature. The DLSPP-WRR-based temperature sensors represent polymer ridge waveguides (~1×1 µm2 in cross section) forming 5-µm-radius rings coupled to straight waveguides fabricated by UV-lithography on a 50-nm-thick gold layer atop a 2.3-µm-thick CYTOP layer covering a Si wafer. A broadband light source is used to characterize the DLSPP-WRR wavelength-dependent transmission in the range of 1480-1600 nm and to select the DLSPP-WRR component for temperature sensing. In- and out-coupling single-mode optical fibers are then glued to the corresponding access (photonic) waveguides made of 10-µm-wide polymer ridges. The sample is heated from 21°C to 46 °C resulting in the transmission change of ~0.7 dB at the operation wavelength of ~1510 nm. The minimum detectable temperature change is estimated to be ~5.1∙10−3 °C for the bandwidth of 1 Hz when using standard commercial optical detectors.
© 2011 OSA
Temperature sensing based on optical techniques is a promising direction in the development of sensor technologies, and remains an area of continuing and intensive research in recent years due to several advantages as compared to other temperature measurement techniques, e.g., high sensitivity, large temperature range, stability and immunity of optical signal to variations and turbulence in the environment . Until now, fiber-optic temperature sensors constitute a major category of the optical temperature sensors, and they mainly employ the principles of fiber Bragg gratings  or surface plasmon resonance (SPR) . In both cases, temperature changes are monitored by detecting variations in the resonance wavelength determined from measured spectra (of Bragg reflection or transmission and SPR excitation). The fiber-optic temperature sensors make use of well-developed fiber-optic techniques, and are very favorable for constructing remote distributed sensing networks, taking advantage of low propagation losses in optical fibers and wavelength division multiplexing techniques. However, all these fiber-optic temperature sensors are bulky and problematic to scale down, and thus can hardly be used as chip-scale temperature sensors. At the same time, recent developments in plasmonic circuitry opened new perspectives for further miniaturization of photonic components based on surface plasmon polariton (SPP) waveguides . With respect to temperature sensing, fiber-coupled dielectric-loaded SPP (DLSPP) waveguide components developed for thermo-optic switching and modulation of radiation  seem rather promising by offering a straightforward way to designing miniature plasmonic temperature sensors.
In this work, we demonstrate the usage of DLSPP waveguide-ring resonators (WRRs)  for sensing the ambient temperature. DLSPP waveguides represent submicron-sized dielectric (polymer) ridges placed atop metal (gold) film surface that confine the SPP modes (supported by the metal-dielectric interface) in the lateral cross section . The dielectric refractive index and thus the DLSPP mode effective index depend on the ambient temperature , whose variations change the phase accumulated by the DLSPP mode circulating in a ring waveguide causing thereby changes in the DLSPP-WRR transmission . Note that the DLSPP waveguide platform allows one to utilize 5-µm-radius WRRs [6, 8] resulting in WRR sensor footprints, which are, for example, two orders of magnitude smaller than that of miniature temperature sensors based on silica/polymer microfiber knot resonators  that are essentially analogous to WRRs. In the following section, we describe the investigated sample and the experimental setup used in this work. Section 3 is then devoted to the optical characterization of the DLSPP-WRR transmission and its response to the temperature variations. Finally, discussion and conclusion are offered in Section 4.
2. Experimental setup and sample
We conducted optical transmission measurements with fiber-coupled DLSPP-WRRs  subjected to different temperatures (Fig. 1 ). As a light source, we used a telecom broadband (1480-1600 nm) source (InPhenix, IPSDM1511c), which was connected to a single-mode fiber (core diameter ~10 µm) with the other end being glued to the sample edge (Fig. 1(a-c)). Ridges (~10 µm in width and ~1.05 µm in height) of poly-methyl-methacrylate (PMMA) polymer were used as access photonic waveguides to guide the radiation to and from the plasmonic section, where funnels (length ~25 µm) narrow the PMMA ridges of access waveguides down to 1-µm-wide ridges of DLSPP-WRR waveguides (Fig. 1(d)). All these dimensions were previously chosen to facilitate the optical coupling between single-mode fibers and DLSPP waveguides . The investigated structures were fabricated by deep UV lithography (wavelength 250 nm) with Süss Microtech MJB4 mask aligner in the vacuum contact mode. A ~1-µm-thick PMMA resist layer was spin-coated on a ~2.3-µm-thick buffer layer of CYTOP supported by a Si wafer substrate (Fig. 1(e)). In the middle of the sample, a 50-nm-thick gold pad supported the funnels and DLSPP-WRR waveguides of the plasmonic section. The DLSPP-WRR structure represented a ~5-µm-radius circular waveguide placed with the gap of ~0.4 µm next to a straight waveguide [5, 6]. An optical (far-field) microscope with an infrared camera was used to visualize radiation scattering at structural junctions (seen as white bright spots in Fig. 1(c)) and facilitate the adjustment of in-coupling fiber . Out-coupling single-mode fiber was carefully positioned from the other side of the sample using exactly the same procedure, since the investigated structure was symmetric with respect to the sample center, and then connected to an optical spectrum analyzer (Agilent 86142B) used for the optical detection. After completing the positional adjustments of in- and out-coupling fibers and controlling the DLSPP-WRR characteristic transmission , both fibers were glued to the sample (Fig. 1(e)). Very small amounts of glue were used, since wetting along the PMMA access waveguide ridges caused by the capillary effect could easily jeopardize the photonic mode confinement resulting in unacceptably large transmission losses. The glue (NOA61 Thorlabs) was cured with UV-light for ≥ 30 min, with the experiment started not sooner than the day after for optimal adhesion.
Sample was resting on a Si-wafer supported by steel shoulders. Placed under the wafer a Peltier element worked as a heating element (Fig. 1(b)). For measuring the sample temperature, a thermocouple was attached the sample about 2-3 mm away from the plasmonic section. No climate chamber was available, so the room temperature and humidity were not controlled. Finally, no dedicated cooling was applied, only radiation and air convection naturally occurred in the laboratory environment contributed to the established temperature. The homogeneity of the temperature distribution across the sample was investigated by using a thermal viewer and found to be within a few degrees.
3. Optical characterization and discussion
LSPP-WRR transmission spectra in the range of 1480-1600 nm were recorded at different temperatures between 21 °C and 46 °C (Fig. 2 ) controlled by the Peltier heating element and measured by a thermocouple as explained above. It was found that the spectral features were more complicated and less pronounced than observed in the first experiments , being generally similar to those obtained with thermo-optic DLSPP components . Similarly to the latter work, we relate this behavior to the circumstance that we used in these experiments slightly wider and taller PMMA ridges than in our first investigations, exceeding the upper limit of single-mode DLSPPW operation . Our choice of these DLSPP waveguide parameters was aimed at realization of the fiber-based end-fire DLSPP excitation . As noted previously , these parameters result in more complex DLSPP-WRR transmission spectra (Fig. 2(a)), since the coupling from a straight waveguide to a ring and back now involves more modes. Note that the sensitivity of DLSPP-WRR-based temperature sensors is expected to be proportional to the slope of its transmission spectrum that constitutes the strongest contribution to the DLSPP-WRR temperature response .
The DLSPP-WRR transmission spectra show variations in the transmission between −41 and −52 dB over a wavelength range of ~45 nm giving a slope of dTr/dλ ≈1.8 µm−1, where dTr is the transmission variation when the wavelength change is dλ. This slope is ~42 times less steep as compared to the first reported experiments , implying that one easily improve the sensitivity by using the appropriate structure parameters. Continuously repeated (over several hours) transmission measurements at the nominal (with the Peltier heating element being switched off) room temperature of 21 °C and 46 °C indicated considerable variations in the transmission level (Figs. 2(a) and (b)). The results of these measurements of long-term stability can be summarized as follows: ΔTr = (0.69 ± 0.18)dB/25°C, which results in a minimum detectable temperature change of dTmin ≈6.4 °C. These relatively large variations of transmission (measured for the same temperature) and thereby low temperature sensitivity originated from various reasons, of which uncontrolled humidity (that can also influence the PMMA refractive index) and temperature changes in the laboratory in the course of our experiment that lasted several hours were probably the most important ones.
We have also conducted relatively quick measurements of the DLSPP-WRR transmission spectra, varying the temperature in steps of 5 °C in less than 10 minutes. It is to be noted that the sample temperature measured with a thermocouple was found to be settling to a constant value after changing the current in the Peltier element during one minute. Fast measurements of the transmission spectra showed monotonous variations (increase or decrease, depending on the wavelength) in the DLSPP-WRR transmission with the temperature (Fig. 2(c)). In this case, the reference transmission spectrum was first obtained at the room temperature of 21 °C and then changes in transmission were recorded at the corresponding temperature that was varied in steps of 5 °C with the Peltier heating element. It is seen that the changes in transmission are proportional to the temperature changes for practically all wavelengths, especially in the wavelength ranges of strong transmission changes (Fig. 2(c)). Furthermore, upon comparison with the transmission spectra shown in Fig. 2(a) one notices that temperature-induced transmission variations are overall proportional to the slope (dTr/dλ) of the wavelength-dependent transmission as expected .
The measured spectra of temperature-induced transmission variations allow one to easily identify the most suitable wavelength for temperature sensing, that in our case was chosen to be 1511 nm (Fig. 2(c)). A linear fit to the change in transmission at the wavelength of 1511 nm as a function of the sample temperature (Fig. 2(d)) resulted in the following sensitivity value: dTr(dB)/dT = 0.023 dB/°C or, in absolute units, dTr/dT ≈2.8∙10−7/°C. The minimal detectable temperature change can now be estimated by considering an experimental setup with a conventional optical detector as follows:
In summary, we have demonstrated fiber-pigtailed miniature (footprint of ~100 µm2) DLSPP-WRR-based temperature sensors suitable for remote control of the ambient temperature. The characterization of temperature sensitivity conducted repeatedly over several hours showed large deviations resulting in poor temperature sensitivity of dTmin ≈6.4 °C. This can be explained with the lack of a climate chamber to perform the experiments under well controlled conditions with respect to both ambient temperature and humidity. Note that, most probably, a significant contribution to the observed variations (Fig. 2(b)) was in fact related to actual changes in the ambient temperature. Temperature induced changes in the transmission measured over short periods of time indicated the possibility of achieving the temperature sensitivity better than 10−2 °C when using a low-power laser and a simple optical detector. The obtained characteristics can significantly be improved by using DLSPP waveguide parameters ensuring the single-mode operation [6, 7]. Further progress is expected when extending the design parameter space by considering race-track  and disk  resonators.
The reported experiments open up a new promising application avenue for fiber-coupled DLSPP-based components, viz., environmental sensing. Recently demonstrated on-chip monitoring of the DLSPP waveguide mode power  adds an exciting possibility of designing integrated plasmonic sensing chips containing temperature sensors along with power monitors (based on integrated Wheatstone bridges  or integrated Seebeck junctions ) for signal detection. This integration can naturally be followed by attaching light sources that are end-fire coupled to access photonic and/or DLSPP waveguides and thereby developing completely autonomous, miniature plasmonic temperature sensors that would preserve the main advantage of conventional optical sensors, viz., the immunity to external electromagnetic signals. Moreover, the very recent development of DLSPP-based waveguide components integrated into a silicon-plasmonic router architecture with 320 Gb/s throughput capabilities  opens an important application avenue for the investigated DLSPP-WRR-based temperature sensors, viz., local temperature control inside silicon-plasmonic router chips since both plasmonic interconnect components  and sensors can be manufactured using the same fabrication technology platform.
This work was financially supported by the Danish Council for Independent Research (FTP-project ANAP, contract No. 09-072949) and by the Regional Council of Bourgogne.
References and links
1. P. R. N. Childs, J. R. Greenwood, and C. A. Long, “Review of temperature measurement,” Rev. Sci. Instrum. 71(8), 2959–2978 (2000). [CrossRef]
2. B. Lee, “Review of the present status of optical fiber sensors,” Opt. Fiber Technol. 9(2), 57–79 (2003). [CrossRef]
3. A. K. Sharma, R. Jha, and B. D. Gupta, “Fiber-optic sensors based on surface plasmon resonance: A comprehensive review,” IEEE Sens. J. 7(8), 1118–1129 (2007). [CrossRef]
4. D. K. Gramotnev and S. I. Bozhevolnyi, “Plasmonics beyond the diffraction limit,” Nat. Photonics 4(2), 83–91 (2010). [CrossRef]
5. J. Gosciniak, S. I. Bozhevolnyi, T. B. Andersen, V. S. Volkov, J. Kjelstrup-Hansen, L. Markey, and A. Dereux, “Thermo-optic control of dielectric-loaded plasmonic waveguide components,” Opt. Express 18(2), 1207–1216 (2010). [CrossRef] [PubMed]
7. T. Holmgaard and S. I. Bozhevolnyi, “Theoretical analysis of dielectric-loaded surface plasmon-polariton waveguides,” Phys. Rev. B 75(24), 245405 (2007). [CrossRef]
8. T. B. Andersen, Z. H. Han, and S. I. Bozhevolnyi, “Compact on-chip temperature sensors based on dielectric-loaded plasmonic waveguide-ring resonators,” Sensors (Basel Switzerland) 11(2), 1992–2000 (2011). [CrossRef]
9. Y. Wu, Y.-J. Rao, Y.-H. Chen, and Y. Gong, “Miniature fiber-optic temperature sensors based on silica/polymer microfiber knot resonators,” Opt. Express 17(20), 18142–18147 (2009). [CrossRef] [PubMed]
10. J. Gosciniak, V. S. Volkov, S. I. Bozhevolnyi, L. Markey, S. Massenot, and A. Dereux, “Fiber-coupled dielectric-loaded plasmonic waveguides,” Opt. Express 18(5), 5314–5319 (2010). [CrossRef] [PubMed]
11. O. Tsilipakos, E. E. Kriezis, and S. I. Bozhevolnyi, “Thermo-optic microring resonator switching elements made of dielectric-loaded plasmonic waveguides,” J. Appl. Phys. 109(7), 073111 (2011). [CrossRef]
12. S. Randhawa, A. V. Krasavin, T. Holmgaard, J. Renger, S. I. Bozhevolnyi, A. V. Zayats, and R. Quidant, “Experimental demonstration of dielectric-loaded plasmonic waveguide disk resonators at telecom wavelengths,” Appl. Phys. Lett. 98(16), 161102 (2011). [CrossRef]
13. A. Kumar, J. Gosciniak, T. B. Andersen, L. Markey, A. Dereux, and S. I. Bozhevolnyi, “Power monitoring in dielectric-loaded surface plasmon-polariton waveguides,” Opt. Express 19(4), 2972–2978 (2011). [CrossRef] [PubMed]
14. J.-C. Weeber, K. Hassan, A. Bouhelier, G. Colas-des-Francs, J. Arocas, L. Markey, and A. Dereux, “Thermo-electric detection of waveguided surface plasmon propagation,” Appl. Phys. Lett. 99(3), 031113 (2011). [CrossRef]
15. S. Papaioannou, K. Vyrsokinos, O. Tsilipakos, A. Pitilakis, K. Hassan, J.-C. Weeber, L. Markey, A. Dereux, S. I. Bozhevolnyi, A. Miliou, E. E. Kriezis, and N. Pleros, “A 320 Gb/s-throughput capable 2×2 silicon-plasmonic router architecture for optical interconnects,” J. Lightwave Technol. 29(21), 3185–3195 (2011). [CrossRef]