Microstructured optical fibers (MOFs) show plasmonic properties after deposition of metal nanoparticles on the surface of their capillaries. A method of enhancing the functionality of such fibers by immobilizing different nanoparticles in the different capillaries of an MOF is described. Silver and gold nanoparticles show well-separated localized surface plasmon resonances (LSPRs). Measurements confirm calculations, according to which both resonance wavelengths shift with changes in the refractive index of the nanoparticles’ immediate environment. Such modified MOFs can be used in LSPR sensing; they may also be used in multiplex detection of bio-analytes, in particular.
© 2012 OSA
MOFs are a special group of optical fibers with a complex inner structure. The materials can vary greatly. Since their invention in 1995 , they have opened up many diverse fields of application. They have been designed to fulfill different tasks. Photonic bandgap fibers are well known for their light guidance in air/gas or fluid over very large interaction lengths [2,3]. Others have been used to create new light sources, such as supercontinuum sources . In so-called Kagome fibers, photocatalytic reactions have been monitored by spectral analysis of the source materials and the reaction product . Further applications include fibers with endless single-mode guidance down to very short wavelengths. MOFs have been additionally modified by filling capillaries with highly diverse materials like polymers , chalcogenide glass , or metals . In this way, properties like dispersion, birefringence, or photonic bandgap guiding can be tuned.
One novel idea is to combine MOFs with defined monolayers of metal nanoparticles. In 2010 we presented a technique for the deposition of nanoparticle monolayers in the capillaries of MOFs . This nanoparticle layer deposition (NLD) technique represents a reliable method for immobilizing gold nanoparticles in an MOF. To our knowledge, this is the only technique that works, even in not-perfectly-circular capillaries, for depositing homogeneous metal monolayers over several meters. These layers remain stable for many years. The achievable nanoparticle density is, with a value of about 450 particles/µm2, quite high; however, the nanoparticles do not form a closed metal layer. Thus, light can excite localized surface plasmon resonance (LSPR) in such layers. The resonance wavelength of the localized surface plasmons is strongly correlated to the metal used, as well as to the size and shape of the nanoparticles . The shift of the resonance wavelengths depends on the properties of the material surrounding the nanoparticles and is, therefore, a useful parameter in sensing. Binding specific capture molecules to the surface of the nanoparticles – a well-established bio-functionalization technique in biosensor and biochip technologies – makes them susceptible to different analytes (e.g., specific biomolecules). Therefore, modified MOFs with well-defined plasmonic layers of metal nanoparticles can be used as versatile nano-cuvettes for the detection of biomolecules in solution.
In this paper, we will present a method which allows selective deposition in single capillaries of the MOF and in which two different kinds of metal nanoparticles are deposited, one in each of the different capillaries. The experiments demonstrate that the chosen nanoparticles have well-separated LSPR wavelengths. By immobilizing different capture molecules on the Au and the Ag nanoparticles multiplex detection of specific biomolecules should be possible. This would make such modified MOFs attractive for efficient biosensing systems.
2. Experimental work
Thirty nm gold (Au) spheres show a typical LSPR at about 530 nm both in solution and immobilized on glass substrates. We have chosen 23 nm silver (Ag) spheres for the second kind of particles with a resonance peak at 400 nm. This should be far enough apart to separate both resonance peaks clearly but still close enough together to be measured with a common spectrometer. Both particles were synthesized based on established wet-chemical processes [10,11]. The particles are mainly spherical with a certain size distribution as revealed by the transmission electron microscope (TEM) images shown in Fig. 1 .
The mean size was determined to be 29 nm ± 1.7 nm for gold and 23 nm ± 1.5 nm for silver particles. For immobilization of the nanoparticles on glass, a chemical modification of the surfaces is necessary. Therefore, functionalized sil(ox)anes form an adhesive, self-assembled monolayer (SAM)  prior to the attachment of nanoparticles.
The MOF produced in house is a three-hole suspended core fiber made of pure silica [Fig. 2(a) ] with hole dimensions of 30 x 40 µm, a core diameter of about 3 µm, and an outer diameter of 125 µm . We used the setup shown in Fig. 2(b) to select single capillaries for sequential NLD processes. Tiny drops of epoxy glue were deposited precisely at the end face of the MOF to seal a single capillary . The glue was sucked into the fiber by capillary forces and finally cured by UV light. The sucking process can be accelerated by applying a vacuum at the opposite end of the fiber. It turns out that a commercially available graphite ferrule (MasCom) achieves the best pressure-tight connection between the MOF and any gas or fluidic system. Its pressure stability reaches up to at least 10 bars. Such graphite ferrules have also been used to connect the MOF to the fluidic system necessary for the NLD process described more in detail in .
After sealing one of the three capillaries, the fiber will be connected to a fluidic system and Au nanoparticles can be deposited by NLD in the two open capillaries. Once the monolayers therein are completed, the fiber is removed from the fluidic system and these two capillaries are sealed as described above. The sealing of these two capillaries must go much deeper than the sealing of the other one [Fig. 2(c)]. This process is controlled via microscope. Then, the fiber is cleaved so that the two modified capillaries stay sealed and protected from following NLD procedure whereas the third capillary is now open. Finally the fiber will be connected again to the fluidic system and the now open capillary can be modified with Ag nanoparticles by a second NLD process.
Calculations  based on the Mie theory predict LSPRs centered at around 400 nm (Ag) and 530 nm (Au) for the nanoparticles used. Extinction measurements have been carried out with a halogen light source (DH2000, Ocean Optics) and a fiber spectrometer (Spectro 320D, Instrument Systems). Any measurement along the core of such plasmonic MOFs with high nanoparticle surface coverage is hampered due to the very strong interaction of the nanoparticles with the guided light. Therefore, the plasmonic MOF is illuminated and measured perpendicular to the fiber axis (Fig. 3 ).
A fiber with a 50 µm core acts like a pinhole for the spectral extinction measurement. It selects that part of the light Imess(λ) which is transmitted through the plasmonic MOF and guides it into the fiber spectrometer. The extinction spectra E(λ) have been calculated by
This unusual setup for measuring extinctions on fibers simplifies the measurement and the preparation in this basic state of investigation because it allows separating optical and fluidic paths from each other. Additionally, there is no need to launch light into the tiny 3 µm core of the MOF.
Using the NLD technique and the method for selective blocking of capillaries described above, we successfully immobilized 29 nm Au and 23 nm Ag nanoparticles selectively in the single capillaries of an MOF homogeneously over a length of half a meter. A simple lateral microscope image [Fig. 4(a) ] confirms this result. A capillary with a Au nanoparticle layer appears purple in transmission due to a lack of green wavelengths (LSPR around 530 nm), whereas a capillary with a Ag nanoparticle layer appears yellowish due to the missing blue wavelengths (LSPR around 400 nm).
The homogeneity of the layers across the entire length of 0.5 m could be confirmed by scanning electron microscope (SEM) images of the Au and Ag nanoparticle monolayers [Figs. 4(b) and 4(c)] taken at both ends of the fiber. The density of the Au and Ag nanoparticles achieved has been determined as 350 nanoparticles/µm2 and 450 nanoparticles/µm2, respectively.
Mie calculations  and experiments suggest that the LSPR of 23 nm Ag particles is at least 10 times more intense than that of 29 nm Au particles. The significant difference in the peak extinctions can be diminished by modifying two capillaries with Au nanoparticles but only one with Ag nanoparticles. This facilitates the analysis of the LSPRs measured. Electron probe microanalysis (EPMA) on the end face of this fiber showed that there were actually Au nanoparticles in two and Ag nanoparticles in one of the three capillaries of the fiber.
Figure 5 shows a comparison between the theoretically calculated (Mie theory) and the transversally measured extinction spectra. The LSPRs are centered at wavelengths typical for the nanoparticles chosen, and the extinction ratio of both peaks is as expected according to the theoretical estimation. In the experimental spectrum, both peaks are broadened and slightly shifted to longer wavelengths, probably due to the variations in the size and shape of the particles. Additionally, they do not act as single particles in a homogeneous surrounding medium, which is used as a basis for the theoretical calculations. The particles are attached on the glass surface, and for a certain number of particles, the distance between them is smaller than their diameter. Both facts can influence the plasmonic properties of the particles. Nevertheless, both LSPRs show the expected positions and are well separated from each other – a main requirement for performing multiplex experiments.
In order to determine the sensitivity of both kinds of nanoparticles, a water-based analyte with a refractive index that changes from 1.33 to 1.43 was pumped successively through the plasmonic MOF, and the spectral shift of the LSPRs was measured. Between measurements, the MOF was cleaned by washing with destilled water. As expected from Mie calculations, the peak positions of the LSPR of Au and Ag nanoparticles move to longer wavelengths if the refractive index of the solution surrounding them is increasing. This effect is demonstrated in Fig. 6(a) . A normal log fit was used to estimate the exact peak positions of both LSPRs, which are obviously not symmetrical. Figure 6(b) outlines all peak positions over the refractive index of the analyte.
The sensitivity of the plasmonic layers can be calculated from the change in the LSPR peak position ΔλLSPR for a defined change in the refractive index Δn of the surrounding medium. One hundred thirty nm/refractive index unit (RIU) for the Ag nanoparticles and 73 nm/RIU for the Au nanoparticles confirm the theoretically estimated values of about 128 nm/RIU for Ag and 71 nm/RIU for Au . A comparison of the results for Au nanoparticles with earlier data  demonstrates the reproducibility of the preparation and utilization of metal nanoparticle layers in MOFs.
The NLD technique enables the production of MOFs with homogeneous monolayers of metal nanoparticles in the capillaries. Such MOFs have plasmonic properties. The technique has been upgraded to prepare layers from different kinds of metal nanoparticles on the inner surface of the single capillaries of an MOF. Layers of 29 nm Au and 23 nm Ag spherical nanoparticles have been deposited homogeneously in the capillaries of one suspended core fiber over a length of 0.5 m. It should be possible to obtain lengths in the meter range, which would allow an effective method of producing hundreds of identical fiber pieces for sensor applications. The typical and well-separated LSPR peaks of these particles have been measured. Changes in the refractive index of the surrounding medium can be detected with 73 nm/RIU for Au and 130 nm/RIU for Ag.
A possible application could be the simultaneous detection of two different biomolecules. To achieve this one type of capture molecules should be sucked into the capillary with Ag nanoparticles to bind to them. In the other two capillaries with Au nanoparticles a second type of capture molecules should be sucked and bound. This sensibilization can be done either immediately after the NLD process in each capillary or on the completed plasmonic MOF using the sealing and cleaving procedure as described above. By this bio-functionalization plasmonic MOFs can serve as a basis for multiplex sensing.
We thank Franka Jahn for the excellent SEM and TEM images and Julia Miteva for the editorial support.
References and links
1. P. St. J. Russell, “Photonic-crystal fibers,” J. Lightwave Technol. 24(12), 4729–4749 (2006). [CrossRef]
2. F. Benabid, G. Bouwmans, J. C. Knight, P. St. J. Russell, and F. Couny, “Ultrahigh efficiency laser wavelength conversion in a gas-filled hollow core photonic crystal fiber by pure stimulated rotational Raman scattering in molecular hydrogen,” Phys. Rev. Lett. 93(12), 123903 (2004). [CrossRef] [PubMed]
3. Y. Huang, Y. Xu, and A. Yariv, “Fabrication of functional microstructured optical fibers through a selective-filling technique,” Appl. Phys. Lett. 85(22), 5182–5184 (2004). [CrossRef]
4. J. K. Ranka, R. S. Windeler, and A. J. Stentz, “Visible continuum generation in air-silica microstructure optical fibers with anomalous dispersion at 800 nm,” Opt. Lett. 25(1), 25–27 (2000). [CrossRef] [PubMed]
5. J. S. Y. Chen, T. G. Euser, N. J. Farrer, P. J. Sadler, M. Scharrer, and P. St. J. Russell, “Photochemistry in photonic crystal fiber nanoreactors,” Chemistry 16(19), 5607–5612 (2010). [PubMed]
6. M. Balakrishnan, R. Spittel, M. Becker, M. Rothhardt, A. Schwuchow, J. Kobelke, K. Schuster, and H. Bartelt, “Polymer-filled silica fibers as a step towards electro-optically tunable fiber devices,” J. Lightwave Technol. 30(12), 1931–1936 (2012). [CrossRef]
7. M. A. Schmidt, N. Granzow, N. Da, M. Peng, L. Wondraczek, and P. St. J. Russell, “All-solid bandgap guiding in tellurite-filled silica photonic crystal fibers,” Opt. Lett. 34(13), 1946–1948 (2009). [CrossRef] [PubMed]
8. R. Spittel, D. Hoh, S. Brückner, A. Schwuchow, K. Schuster, J. Kobelke, and H. Bartelt, “Selective filling of metals into photonic crystal fibers,” Proc. SPIE 7946, 79460Z, 79460Z-8 (2011). [CrossRef]
9. A. Csaki, F. Jahn, I. Latka, T. Henkel, D. Malsch, T. Schneider, K. Schröder, K. Schuster, A. Schwuchow, R. Spittel, D. Zopf, and W. Fritzsche, “Nanoparticle layer deposition for plasmonic tuning of microstructured optical fibers,” Small 6(22), 2584–2589 (2010). [CrossRef] [PubMed]
10. J. Turkevich, P. C. Stevenson, and J. Hillier, “A study of the nucleation and growth processes in the synthesis of colloidal gold,” Discuss. Faraday Soc. 11, 55–75 (1951). [CrossRef]
11. G. Frens, “Controlled nucleation for the regulation of the particle size in monodisperse gold suspensions,” Nat. Phys. Sci. (Lond.) 241, 20–22 (1973).
13. J. Kobelke, K. Schuster, A. Schwuchow, D. Litzkendorf, R. Spittel, J. Kirchhof, and H. Bartelt, “Fabrication and Characterization of Special Microstructured Fibers,” Proc. SPIE 8001, 80011H (2011). [CrossRef]
14. M. Zobel, “LSPR - Refraktometrie in selektiv innenbeschichteten mikrostrukturierten optischen Fasern,” Bachelor’s thesis (2011).
16. A. Zaheer, “Determination of refractive index sensitivity of ensemble of Au and Ag nanoparticles of different shapes and sizes, on experimental and simulation basis,” Master’s thesis (2012).