Open Access
Add to BibSonomy
Abstract
The aim of this work is to optically detect the condensation of acetone vapor on an aluminum plate cooled down in a two-phase environment (liquid/vapor). Sub-micron period aluminum based diffraction gratings with appropriate properties, exhibiting a highly sensitive plasmonic response, were successfully used for condensation experiments. A shift in the plasmonic wavelength resonance has been measured when acetone condensation on the aluminum surface takes place due to a change of the surrounding medium close to the surface, demonstrating that the surface modification occurs at the very beginning of the condensation phenomenon. This paper presents important steps in comprehending the incipience of condensate droplet and frost nucleation (since both mechanisms are similar) and thus to control the phenomenon by using an optimized engineered surface.
© 2017 Optical Society of America
1. Introduction
Today, metallic periodic structures are increasingly being used in optics as biological and chemical sensors [1]. They provide strong localization of the electric field at their surface by means of electrons oscillating collectively with the incident electric field. These excitations are known as surface plasmon resonance (SPR). The presence of species on the surface induces a modification of the local refractive index leading to a change of the plasmon resonance frequency because of interactions happening in the evanescent electric field [2]. Thus, SPR sensors are able to detect changes on the surface in real-time [3].
Different kinds of experimental set-ups were used to detect a change of the refractive index such as the attenuated total reflection (ATR) method, optical waveguide-based SPR systems or grating coupler-based SPR system [4, 5]. Metallic grating structures permit a direct excitation of a plasmonic wave. The optical signal (reflection or transmission) depends on the period, the depth, the grating profile [6], as well as on the polarization, the angle and the wavelength of the incident light: it reaches maximum amplitude of absorption (deep in the 0th reflected order) for TM polarized incident light absorption on 1D grating structures [7] or in 2D grating for un-polarized incident wave [8]. There are two principal ways to manage measurements with such sensors: by varying the angle of the incident wave at a fixed wavelength or by a change of the wavelength of the incident light keeping a fixed angle. In this second case, for a given angle, the plasmon resonance wavelength can be measured experimentally by analyzing the reflection or absorption spectrum [9].
Many types of metals are available for their interesting plasmonic properties, such as gold, and copper for the visible and the near infrared wavelength range. Aluminum is also a good candidate, exhibiting strong enhanced local fields [10]. Furthermore, aluminum is one of the most promising materials that can be used in the thermal domain, because of its high thermal conductivity and its lightness.
In parallel, many researches have recently been using nano -coatings in the thermal domain to obtain a super hydrophobic or super hydrophilic surface [11]. Such a coating can for example prevent or increase the phenomenon of condensation, which is a challenging issue in many applications such as energy saving (for example in air heat pumps), water desalination or chemical engineering, etc [12]. As an application oriented example, the energy efficiency of outdoor air heat pumps is significantly decreasing with the air temperature because of frost formation on the evaporator. Controlling or detecting the incipience of frost nucleation as well as defrosting is a very challenging industrial objective. The optimization of these surfaces could be improved if it was possible to measure the early formation of droplets, which is a fundamental issue still difficult and complex to understand [13–16].
Condensation on such surfaces can be estimated by measuring the overlap of surface by the liquid molecules. Super-hydrophobic surfaces with “egg box” shape were imaged by optical microscopy after the water condensation phenomenon in order to deduce the spreading of the liquid on the surface. The authors measured only the contact angle and the size of the droplets to detect their coalescence [17]. Another way to study the condensation process at an earlier stage is to image the formation of droplets using environmental scanning electron microscopy (SEM). Indeed, the SEM images made it possible to study the condensation for example on super-hydrophobic surfaces (knife-like CuO structures) [18] or on hydrophobic - hydrophilic hybrid surfaces (parallel lines etched by focused ion beam) [19]. However, SEM technology is very difficult to implement at the very beginning of the phenomena.
Since SPR is well adapted to detect changes within the surrounding medium at a metallic surface, this method has been investigated here to detect the condensation phenomenon in a thermal chamber containing a two-phase working fluid (acetone). A hydrophilic surface was used and the plasmon is excited at the dielectric-aluminum interface by the incident light coupled by the periodic structure. This 1D structure illuminated in TM polarization is more efficient for accurate plasmon resonance detection than 2D structures [20, 21]. The detection of condensation on the aluminum sample is realized using a grating fabricated by nanoimprint [22] and the optical response has been investigated both experimentally and theoretically.
First, a simulation is performed to optimize the parameters of the structure in air or acetone vapor (n = 1) in order to find a strong and narrow resonance signal. Then, fabrication of the grating and the set-up for plasmon detection are described before the presentation of the experimental results. Finally, a study and an analysis of the resonance shift due to the condensation of acetone are conducted by comparing the experimental to simulated results. Previous experiments (not presented in this paper) have been conducted by the authors to verify that the metallic corrugation doesn’t influence the condensation phenomenon since the grating profile is very smooth compared to the first droplets size.
2. Modeling
2.1 Grating design and resonance modeling
The ability of condensation detection depends on the amplitude and on the narrowness of the plasmon resonance, which is linked in particular to the profile (square or sinus), to the period (Λ) and to the depth (d) of the metallic grating. The studied structure is shown in Fig. 1(a). The diffraction grating is composed of a photoresist material on an aluminum substrate. Then, an aluminum film of thickness (e) covers the entire sample in order to create a structured metallic surface without any etching process. Finally, an acetone layer of thickness (p) will appear progressively upon the structure during the condensation/frost phenomenon while the substrate is cooled. A sinusoidal profile was chosen, since it corresponds to the best plasmon signature compared to other types of grating profiles.
Fig. 1 (a) 1D aluminum grating at incident angle θ = 20°, Λ and d are the period and the depth of the grating, e and p are the thicknesses of the aluminum and of the condensate layer respectively; Simulation of TM reflectance in vapor phase (p = 0) for: (b) two periods Λ = 760 nm and Λ = 1000 nm (e = 50 nm, d = 55 nm); (c) different grating depths (Λ = 760 nm, e = 50 nm) and (d) different aluminum thicknesses (Λ = 760 nm, d = 55 nm).
The structure is illuminated in air by a collimated polychromatic light at an incident angle of θ = 20° under TM polarization and the studied signal is the grating’s 0th reflected order.
A simulation using a commercial program (MC grating) based on the Chandezon method [23, 24] is performed to choose the parameters of such a 1D structure in order to obtain the best plasmonic resonance for the detection of acetone condensation.
The influence of the grating characteristics and the aluminum layer thickness on the amplitude of the resonance is first studied in pure vapour phase (p = 0). For a grating depth d = 55 nm and an aluminum thickness e = 50 nm, Fig. 1(b) represents a comparison of reflectance spectra for two different periods of the structure corresponding to different resonance wavelength. The period of Λ = 760 nm leads to a main resonance at the wavelength λR = 520 nm, which is more important than the one for Λ = 1000 nm at λR = 669 nm. This result leads to choose the period Λ = 760 nm with the same aluminum thickness (e = 50 nm). With these parameters, Fig. 1(c) shows that the grating depth of d = 55 nm induces the most important resonances with regard to their amplitude. Finally, with Λ = 760 nm and d = 55 nm, as it is illustrated in Fig. 1(d), the thickness of the aluminum film can also have a great influence on the resonance amplitude. Indeed, for a thin Al layer (e < 20 nm), a part of the incident signal will be transmitted into the photoresist and will not contribute to the plasmon effect (for e = 10 nm the resonance amplitude is 38% instead of 63% for e > 40 nm).
According to this study, a sinusoidal grating has been designed and optimized, leading to the following parameters: period Λ = 760 nm, depth d = 55 nm and aluminum thickness e = 50 nm (> 40 nm).
2.2 Electromagnetic field distribution
For the optimized structure, Fig. 2 illustrates the electric field distribution for the resonance wavelength (λ = 520 nm) and a TM polarized incident beam. It shows the xOz plane, with the x-axis being oriented along the surface, perpendicular to the grating lines and normalized to the grating period Λ, and the z-axis being oriented perpendicular to the surface. The light propagates from the top (vapor p = 0) to the bottom (Al). The output electric field is also normalized with respect to the incident excitation field.
Fig. 2 Simulation of electromagnetic field around an Al 1D grating with vapor phase as cover with the parameters: Λ = 760 nm, d = 55 nm, e = 50 nm, εd = 1, εm’ = - 38.5 + i11.
One can observe that the electric field of the plasmon mode extends about one wavelength into the air/vapor medium, compared to a very shallow penetration of the metal of some nm only. The skin depth zm into the metal for the plasmon resonance wavelength and the penetration depth zd of the electric field into the dielectric medium (vapor phase) are given by Eq. (1) and Eq. (2), respectively [25]:
εm and εd are the real parts of the permittivity of the metal and of the dielectric medium. With a skin depth for the metal of only zm = 13.1 nm, a thin layer of aluminum (50 nm) is sufficient to avoid the propagation of the light into the photoresist and to assure a good plasmonic effect at the interface air/Al, as has already been demonstrated in the simulation of section 2.1. The electric field has its maximal amplitude in the air at z = 15 nm distance from the air/Al interface, and subsequently decreases exponentially over the distance zm. Consequently, a change at the structure surface due to the condensation phenomenon (formation of liquid phase acetone) would significantly affect the electric field and could be detected by SPR.
Therefore, to verify the ability to detect a change at the interface vapor/Al, the chosen structure was realized as presented in the next section.
3. Grating fabrication and SPR experimental set-up
3.1 Fabrication of aluminum grating
The technological process to obtain the desired metallic 1D Al grating was nanoimprint. Figure 3 shows a schematic diagram of the complete fabrication process of the structure.
Fig. 3 Fabrication of 1D grating of Al, (a) deposition of resist, (b) resist embossing, (c) grating structures and (d) deposition of aluminum on resist.
A photoresist adapted for the subsequent imprint process was first deposited on an 80 mm diameter aluminum substrate by spin-coating (Fig. 3(a)). The structures were then imprinted into the resist using a PDMS grating mold with a period Λ = 760 nm and a depth d = 55 nm (Fig. 3(b)). The mold fabrication was done via laser interference lithography, resulting in a sinus profile grating, which was subsequently copied into PDMS to be used as a mold.
The size of the grating zone is 25 mm square. A 10 nm thin layer of resist remains as residual layer below the grating structure on the surface below the grating (Fig. 3(c)). The metallic structure that creates the plasmonic properties is realized in the form of an aluminum deposition of thickness e = 50 nm on the microstructured photoresist, which was done by magnetron sputtering (Fig. 3(d)). The aluminum layer covers the whole surface, therefore the resist underneath the metallic layer could not be removed, which is however not problematic since the presence of resist does not influence the optical or thermal properties of the substrate: the metallic layer is simply thick enough to be seen as an infinite layer. This type of “encapsulated” sample is resistant in acetone.
The image in Fig. 4 obtained by Atomic Force Microscopy (AFM) represents the 1D grating structure after nanoimprint and aluminum deposition, showing a quasi-sinusoidal profile, with a period (Λ = 760 nm) and the depth (d = 55 nm) of the grating close enough to the desired optimized structure (section 2).
In the following section, this type of structure has been used in a reflectance measurement set-up as a sensor to detect the condensation/frost of acetone at the metallic surface.
3.2 Experimental Condensation set-up
The experimental set-up developed to detect the condensation phenomenon is presented in Fig. 5. The grating was set up vertically in a hermetically sealed reservoir filled with a two-phase working fluid. The reservoir is preliminary emptied and then partially filled with acetone, with the liquid reaching up to the bottom of the sample (Fig. 5(a)). The temperature of the reservoir is controlled and therefore also its pressure in the two-phase conditions. Prior to performing the plasmon experiments, the vapor supply of acetone was brought to a boil, and the test chamber was connected to a vacuum pump to eliminate the non-condensable gases. The plasmon resonance was measured by recording the wavelength spectrum during the condensation phenomenon every min according to the speed of the cool down process of the chamber, which has been performed within several min. A two-mm diameter collimated beam coming from a fiber connected to a halogen source illuminates the grating on an area of 50 mm2 with an incidence angle of θ = 20° through a transparent window of the cooled chamber (Fig. 5(b)). The polarizer is oriented to create TM polarized light. The reflected light is focalized into an optical fiber, which is connected to a spectrometer and analyzed with the spectrometer software.
Fig. 5 (a) Cooling chamber configuration and (b) experimental set-up for reflectance measurements during condensation.
4. Experimental results and discussion
4.1 Measurements
The chamber was first maintained at 25°C, then the fluid was slightly heated. A temperature difference exists between the vapor and the sample, enabling the condensation phenomenon to start. The measured reflectance spectra were plotted as a function of the elapsed time in minutes. This was done to estimate the time needed for the incipience of the frost nucleus and formation of liquid phase acetone (Fig. 6(a)) at the metallic surface. Measurements were performed in the center of the sample and normalized to the reflectance of unstructured aluminum at the edge of the same sample. The absorption dip position of each reflectance curve, which corresponds to the plasmonic resonance wavelength, is represented in Fig. 6(b) as a function of time: the condensation starts in a vapor surrounding medium (n = 1) with a corresponding resonance wavelength λR = 559 nm, with the condensation slowly starting until the 3 minutes mark. Then, the phenomenon becomes faster and the resonance shifts approximately 30 nm towards longer wavelengths within about 90 s. After that, the resonance wavelength stays constant around λR = 611 nm until 5 min 30 s. At 6 minutes, the value of the resonance wavelength suddenly jumps up to 800 nm, after which it increases only slightly to 810 nm over the next minute and finally remains constant until the end of the experiment. The total shift of the resonance wavelength ΔλR/vapor-acetone = 250 nm corresponds to the vapor/liquid phase change of the acetone. Figure 6(a) shows the condensation phenomenon in the wavelength spectrum: The resonance dip wavelength position increases slowly at the beginning (T0 to T0 + 240 s), which corresponds to the first nucleus (small droplets) of either liquid or ice forming, before the complete acetone condensation at the surface suddenly shifts the resonance wavelength to ~800 nm (T0 + 360 s).
Fig. 6 Experimental results: (a) reflectance on the middle of the sample normalized to the reflectance of unstructured Al and (b) plot of resonance wavelengths of each reflectance curves as a function of time.
In order to understand the experimental results (Fig. 6), they were compared to simulations, in particular for the extreme cases of the surrounding medium: the pure vapor phase (n = 1, t = 0) and the liquid phase of acetone (n = 1.36, t = 9 min).
As shown in Fig. 7, the main effect, i.e. a shift of the resonance wavelength by approximately 250 nm between the two configurations, can be reproduced very well in the simulation. For the liquid phase (n = 1.36, t = 9 min), a resonance wavelengths difference of ΔλR/liquid = 19 nm between simulation (λR = 791nm) and experiment (λR = 810 nm) is observed, probably caused mainly by an uncertainty in the incidence angle, since already a shift of Δθ = 1.4° (θ = 18.6° instead of 20°) induces theoretically a change of the resonance wavelength of ΔλR/liquid = 19 nm. Furthermore, in the simulation, the resonance amplitude is greater and the resonance width is narrower than in the experiment. This difference can be explained by incidence angle and the deviation of the experimental grating depth from the optimized value (Fig. 4). Additionally it can also be explained by its variation over the grating surface, as well as by absorption losses of the plasmon wave due to grating defects.
Fig. 7 Comparison of peak resonance: (a) experimental case of grating realized by nanoimprint and (b) simulation with θ = 20°, Λ = 760 nm, d = 55 nm, e = 50 nm and p = 0 nm.
For the vapor phase of acetone (n = 1, t = 0 min of Fig. 6(b)), the difference in the resonance wavelengths ΔλR/vapor = 41 nm between experiment (λR = 559 nm) and simulation (λR = 520 nm) can be explained in the same way as for the case of the liquid phase by an angle error of Δθ = 1.4° but also by the chosen refractive index (n = 1) while molecules of acetone could be still present during vapor phase (ΔλR/vapor = 0 if Δn = 0.05 and θ = 18.6°).
The mentioned deviations create a resonance wavelength shift between the vapor phase and the liquid acetone phase of ΔλR/vapor-acetone = 251 nm in the experiment instead of ΔλR/vapor-acetone = 275 nm in the simulation.
However, the most interesting conclusion from those preliminary results concerns the reflectance curves and the sensitivity of the SPR to the change between the two phases, that is the curves from T0 to T0 + 240 s (Fig. 6(a)), which are interesting for the scientific community of frost and condensation phenomena.
In between these two phases (vapor/liquid), the condensation phenomenon is more complex to analyze. The following section aims to describe the evolution of the wavelength resonance during the condensation phenomenon using several assumptions of the evolution of the refractive index, or at least the effective index representing the surrounding medium.
4.2 SPR modeling during condensation phenomenon
The surface used in our experiments is hydrophilic according to the measured contact angle of 15° using acetone on the structured Al surface. Different assumptions were proposed to study the process during the change of the surrounding medium at the surface grating, explaining the first resonance spectra before the complete condensation at the metallic surface. It is a very complex physical phenomenon a lot of questions remain. Assumptions concern the incipience of micro acetone droplets at the surface, according to the scientist and the authors of the CETHIL institute. Since it is very difficult to simulate droplets of acetone growing at the surface, the authors considered a non-homogenous medium at the surface. The first one is the growth of a homogeneous layer of acetone on the whole surface with a constant refractive index corresponding to those of acetone. However, due to gravity, droplets roll off the surface and they sweep other droplets to clean the surface for re-nucleation. The second one is the formation of an acetone layer in the grooves of the 1D aluminum structures until the acetone totally fills the grooves, creating a homogenous layer on the surface. In both cases, the resonance wavelength increases over time until it reaches the condensation threshold. In reality, the formation of the acetone layer on the surface is too complicated and inhomogeneous to be rigorously simulated, but there is a relation between the evolution of the thickness of acetone layer and the shift of the SPR dip. Therefore, the authors chose to simulate for the angle of θ = 20° the change of the resonance wavelength by varying the thickness (p) of the acetone layer deposited in the grooves until it entirely covers the grating, as indicated in Fig. 8(b).
Fig. 8 (a) Experimental and simulated peak resonance position of Al 1D grating realized by nanoimprint with θ = 20°, Λ = 760 nm, d = 55 nm and e = 50 nm and with different acetone layer thicknesses and (b) The three steps of acetone condensation corresponding to the three parts of curve (a).
The resulting simulated resonance wavelengths for a timescale up to the shift of 60 nm visible in the first part of Fig. 6(b) (0 < t < 5 min 30 s which corresponds to 559 nm < λR < 611 nm) are plotted in Fig. 8(a), as well as the experimental results during condensation from Fig. 6(b). To simulate the condensation phenomenon, three different steps were considered. The first one is the deposition of acetone in the grooves with a thickness varying from p = 5 nm to p = 30 nm (Fig. 8(b) part 1), the second one is the variation from p = 50 nm to p = 55 nm (Fig. 8(b) part 2), and the third one is the formation of a homogeneous acetone layer on the surface with completely filled grooves (Fig. 8(b) part 3)). From these simulations, the different steps of the condensation phenomenon can be predicted as a function of the elapsed time.
After the third step, the condensate acetone layer thickness (15 nm) above the grating is too high and the electric field does not “see” anymore the change on the metal surface. Above this value, the electric field is no more sensitive to any increase of acetone thickness and the SPR jumps suddenly to a value of λR around 800 nm.
Further analyses of the measurements may provide more information on the condensation mechanism itself but the various steps shown in Fig. 8 make it possible to evaluate the time required for the formation of the condensation as well as the different phases occurring during the incipience of the first nucleus of condensation.
5. Conclusion
In this study, a new technique for condensation/frost detection was presented. A plasmonic submicron structure was used as sensor. The structured surface was fabricated by nanoimprint and investigated using AFM characterization showing a sinusoidal 1D grating profile. The reflected spectrum was studied during the condensation process happening in the two-phase chamber. Using the experimental result obtained by SPR, the condensation process was modeled assuming a variation of the thickness of the acetone condensate layer. This technique can be used for real time analysis of the condensation detection, especially considering a temperature controlled environment which influences the steps of condensation before the incipience of first frost/condensation nucleus. The method is very sensitive to refractive index changes at the surface and thus well adapted to the condensation mechanism and can detect it at a very early stage. Further theoretical and experimental work need to be developed to analyze the physics of the nucleation process at the nanoscale with the condensation and frost scientist community. The topics of frost and condensation mechanism is currently investigated to better understand the incipience of the first nucleus or droplet formation at the surface. Optics with SPR has been used for the first time to observe the phenomenon at the very beginning. It is a very promising approach according to the scientist community of condensation/frost field.
Funding
Agence Nationale pour la Recherche ANR-12-SEED-0003 project.
Acknowledgments
The authors would like to thanks HEF for aluminum deposition and the company SILSEF for the nanoimprint grating fabrication.
References and links
1. J. Homola, “Surface plasmon resonance sensors for detection of chemical and biological species,” Chem. Rev. 108(2), 462–493 (2008). [CrossRef] [PubMed]
2. J. A. Porto, F. J. Garcta-Vidal, and J. B. Pendry, “Transmission resonances on metallic gratings with very narrow slits,” Phys. Rev. Lett. 83(14), 2845–2848 (1999). [CrossRef]
3. S. Jiang, Y. Peng, B. Ning, J. Bai, Y. Liu, N. Zhang, and Z. Gao, “Surface plasmon resonance sensor based on molecularly imprinted polymer film for detection of histamine,” Sens. Actuators B Chem. 221, 15–21 (2015). [CrossRef]
4. T. Lopez-Rios, F. Abelès, and G. Vuye, “Investigation of metallic surface layers on metals by surface plasmon ATR spectroscopy,” J. Phys. Lett. 40(14), 343–346 (1979). [CrossRef]
5. M. Vala, K. Chadt, M. Piliarik, and J. Homola, “High-performance compact SPR sensor for multi-analyte sensing,” Sens. Actuators B Chem. 148(2), 544–549 (2010). [CrossRef]
6. A. Arriola, A. Rodriguez, N. Perez, T. Tavera, M. J. Withford, A. Fuerbach, and S. M. Olaizola, “Fabrication of high quality sub-micron Au gratings over large areas with pulsed laser interference lithography for SPR sensors,” Opt. Mater. Express 2(11), 1571–1579 (2012). [CrossRef]
7. Y. Jourlin, S. Tonchev, A. V. Tishchenko, C. Pedri, C. Veillas, O. Parriaux, A. Last, and Y. Lacroute, “Spatially and polarization resolved plasmon mediated transmission through continuous metal films,” Opt. Express 17(14), 12155–12166 (2009). [CrossRef] [PubMed]
8. Y. Qu, Q. Li, H. Gong, K. Du, S. Bai, D. Zhao, H. Ye, and M. Qiu, “Spatially and spectrally resolved narrowband optical absorber based on 2D grating nanostructures on metallic films,” Adv. Opt. Mat. 4(11), 480–486 (2016). [CrossRef]
9. J. Homola, S. S. Yee, and G. Gauglitz, “Surface plasmon resonance sensors: review,” Sens. Actuators B Chem. 54(1–2), 3–15 (1999). [CrossRef]
10. M. W. Knight, L. Liu, Y. Wang, L. Brown, S. Mukherjee, N. S. King, H. O. Everitt, P. Nordlander, and N. J. Halas, “Aluminum plasmonic nanoantennas,” Nano Lett. 12(11), 6000–6004 (2012). [CrossRef] [PubMed]
11. B. Mondal, M. Mac Giolla Eain, Q. Xu, V. M. Egan, J. Punch, and A. M. Lyons, “Design and fabrication of a hybrid Superhydrophobic-hydrophilic surface that exhibits stable dropwise condensation,” ACS Appl. Mater. Interfaces 7(42), 23575–23588 (2015). [CrossRef] [PubMed]
12. F. C. Cebeci, Z. Wu, L. Zhai, R. E. Cohen, and M. F. Rubner, “Nanoporosity-driven superhydrophilicity: A means to create multifunctional antifogging coatings,” Langmuir 22(6), 2856–2862 (2006). [CrossRef] [PubMed]
13. N. A. Patankar, “Supernucleating surfaces for nucleate boiling and dropwise condensation heat transfer,” Soft Matter 6(8), 1613–1620 (2010). [CrossRef]
14. S. Khandekar and K. Muralidhar, Dropwise Condensation on Inclined Textured Surfaces (Springer, 2014).
15. X. Liu and P. Cheng, “Dropwise condensation theory revisited part II. Droplet nucleation density and condensation heat flux,” Int. J. Heat Mass Transfer 83, 842–849 (2015). [CrossRef]
16. X. Liu and P. Cheng, “Dropwise condensation theory revisited part I. Droplet nucleation radius,” Int. J. Heat Mass Transfer 83, 833–841 (2015). [CrossRef]
17. R. D. Narhe and D. A. Beysens, “Water condensation on a super hydrophobic spike surface,” Europhys. Lett. 75(1), 98–104 (2006). [CrossRef]
18. N. Miljkovic, R. Enright, Y. Nam, K. Lopez, N. Dou, J. Sack, and E. N. Wang, “Jumping-Droplet-Enhanced condensation on scalable superhydrophobic nanostructured surfaces,” Nano Lett. 13(1), 179–187 (2013). [CrossRef] [PubMed]
19. Y. Yamada, T. Ikuta, T. Nishiyama, K. Takahashi, and Y. Takata, “Droplet nucleation on a well-defined hydrophilic-hydrophobic surface of 10 nm order resolution,” Langmuir 30(48), 14532–14537 (2014). [CrossRef] [PubMed]
20. T. Numata, Y. Otani, and N. Umeda, “Optical dew sensor using surface plasmon resonance of periodic Ag nanostructure,” Jpn. J. Appl. Phys. 45(30), 810–813 (2006). [CrossRef]
21. H. Nagasaki, S. Kaneko, K. Iwami, and N. Umeda, “Localised surface plasmon resonance dew sensor for use under low humidity conditions,” Jpn. J. Appl. Phys. 51(2), 1–4 (2012).
22. A. Boltasseva, “Plasmonic components fabrication via nanoimprint,” J. Opt. A, Pure Appl. Opt. 11(11), 4801906 (2009). [CrossRef]
23. N. M. Lyndin, (2014), http://mcgrating.com.
24. Y. Jourlin, S. Tonchev, A. V. Tishchenko, and O. Parriaux, “Sharp plasmon-mediated resonant reflection from an undulated metal layer,” IEEE Photonics J. 6(5), 1–6 (2014). [CrossRef]
25. T. V. Shahbazyan and M. I. Stockman, Plasmonics: theory and applications (Springer, 2014).
