Functionalization of Soft Glass Microstructured Optical Fibers via Silane Coupling Agents

(1)

A Fundamental Study Into the Surface

Functionalization of Soft Glass Microstructured Optical Fibers via Silane Coupling Agents

John E. Debs, Heike Ebendorff-Heidepriem, Jamie S. Quinton, and Tanya M. Monro

Abstract—A method for the functionalization of surfaces within soft glass microstructured optical fibers has been developed, using self assembled monolayers (SAMs) of silane coupling agents. We demonstrate the use of measurements of the fiber capillary fill rate, as a positive test for a functionalized internal surface. A simple theoretical model is used for comparison with measured fill rates. During this work, adsorption kinetics for SAMs of octade- cyltrichlororsilane onto lead silicate glass has been investigated.

This work is a critically important first step for a plethora of ap- plications in biophotonics, chemical fiber sensing as well offering promise for protecting fiber glass from degradation.

Index Terms—Microstructured fibers, soft glass, surface func- tionalization.

I. INTRODUCTION

S

INCE the first demonstration of a microstructured optical fiber (MOF) in 1996 [1], research in the field has developed into a vast and rich range of applications, producing ever increasing and attractive photonic devices outside the realm of fiber communications. The next generation of MOFs has enabled the realization of chemical fiber sensors, MOF lasers, and fibers for proteomics and biophotonics [2]–[5] to name a few.

This paper aims to provide the foundation for a new avenue for MOF technology; namely the surface modification of the internal surfaces within microstructured optical fibers. This approach enables the characteristics of existing fibers to be im- proved, and provides new functionality for fiber applications.

The degradation of glass in the presence of water is well rec- ognized [6], [7], and for the case of MOFs an issue that is of

Manuscript received May 07, 2008; revised July 30, 2008. Current version published April 17, 2009. This work was supported in part by the Australian Research Council under DP0665486. The Centre of Expertise in Photonics ac- knowledges support from the DSTO, Australia and the South Australian State Government.

J. E. Debs was with Smart Surface Structures Group, School of Chemistry Physics and Earth Sciences, Flinders University, Adelaide SA 5001, Australia.

He is now with the Centre of Excellence for Quantum Atom Optics, Department of Physics, The Australian National University, Canberra ACT 0200, Australia (e-mail: [email protected]).

H. Ebendorff-Heidepriem and T. M. Monro are with the Centre of Exper- tise in Photonics, School of Chemistry and Physics, University of Adelaide, Adelaide SA 5005, Australia (e-mail: [email protected]; tanya.

[email protected]).

J. S. Quinton is with Smart Surface Structures Group, School of Chemistry Physics and Earth Sciences, Flinders University, Adelaide SA 5001, Australia (e-mail: [email protected]).

Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/JLT.2008.2004884

practical concern is that the introduction of holes within the fiber cross section can accelerate this degradation [8], reducing the lifetime of the fiber. Silane coupling agents (SCAs) have been applied to conventional fibers both as a protective layer, and as an adhesive promoter for the application of protective polymer coatings [9]. Octadecyltrichlorosilane (OTS) is known to form closely packed, self-assembled monolayers on hydrated oxide surfaces (typically silica, or silica-like surfaces) [10]–[17] Such monolayers render the surface hydrophobic due to the long, out- ward oriented alkyl chain, which act to passivate it and thus chemically protect the bulk glass from further water adsorption/absorption [15].

Such coupling agents can also be used to provide new surface functionalities, for example, by providing a functional group for chemically specific molecular binding. In contrast to polymers, proteins cannot bond directly to a glass surface. The general surface functionalization approach presented here has provided a route to achieve protein immobilization at the surface of a glass core; work which was recently presented in [18]. This, in turn, paves the way towards selective detection of proteins using glass MOFs, as has been demonstrated for polymer MOFs [4]. This would be a huge step forward in a number of areas including fiber sensing, proteomics, and biophotonics. In addition, mod- eling of fluorescence-based fiber sensors has demonstrated that immobilization of molecules to be detected at the fiber core will enhance the efficiency with which the fluorescence is captured in the fiber core [19].

We are interested in soft glass MOFs for several reasons. Soft glasses offer optical and thermal properties that are not available with silica. For example, soft glasses can transmit light in the midinfrared spectral region and exhibit high nonlinearity. They have lower glass softening points than silica, enabling preform fabrication via extrusion through complex stainless steel dies, which in turn, has enabled a wealth of new and exciting MOF hole geometries [2], [20].

We demonstrate that films of self-assembled monolayers (SAMs) of octadecyltrichlorosilane (OTS) can be successfully applied to the internal surfaces within a lead silicate MOF. This is achieved by measuring the resultant change in the surface tension within the holes of the MOF, as manifested in modifica- tions to the rate at which fluids can fill the fiber holes. We also provide evidence for monolayer formation on the lead silicate (soft) glass used via FTIR spectroscopy, and contact angle measurements of the film. To the best of our knowledge, there has been no literature on the application of SCA monolayers to soft glasses. This provides a proof of principle that MOFs can be surface functionalized using self assembled layers of SCAs.

(2)

Fig. 1. SEM image of the SF57 suspended core MOF used in this work. The core has a diameter of1.3m and the struts a length of6m.

Fig. 2. Effective mode area as a function of core diameter, calculated for an air suspended rod of SF57 [22].

II. BACKGROUND ANDEXPERIMENTAL

A. Fiber and Glass Choice

The glass selection for the fiber used in this study is a lead silicate glass purchased from Schott Glass Company referred to as SF57. This glass is generally of interest in nonlinear fiber applications, and has been used to produce highly nonlinear MOFs [21], [22]. Furthermore, it consists of approximately a 1:1 SiO :PbO mixture (measured via energy dispersive x-ray analysis), and as an oxide glass, we expect it to display similar surface chemistry to SiO (silica) for our application. To investigate the behavior and kinetics of silane SAMs on SF57, flat 3 mm thick plates were first polished to , and then cut into 4

4 mm samples with a diamond saw.

Fig. 1 shows an SEM image of the suspended core MOF geometry selected for functionalization within this study. In this geometry, the small triangular core in the centre of the fiber acts as a nanowire that is surrounded by three large air holes, and isolated from the external environment by three long, fine supporting struts. There are two regimes of confinement of the guided mode (Fig. 2): for core diameters larger than or comparable to the guided wavelength ( m for SF57), tighter confinement results from a smaller core. Whereas for diameters significantly smaller than the guided wavelength ( m for SF57), decreasing core sizes have less confined modes, resulting in a higher optical power in the air filled region. The first regime leads to high intensities in the core, making the fiber geometry suitable for nonlinear optical applications [21], [22].

While for the latter regime, a higher intensity in the cladding fa- vors sensing applications [5], with a high air filling fraction also enabling a large volume of fluid to be introduced.

B. OTS Solution Preparation

OTS CH CH SiCl was purchased from Sigma- Aldrich at 90% purity (the remaining 10% is composed of long chain alkanes), and a solution of OTS in Hexadecane (Sigma 99% purity) was made to 2 mM concentration. The hexadecane was dehydrated by storing it with molecular sieves. Formation of monolayers is via the hydrolysis of the chlorine head groups with surface water to form silanols, followed by condensation reactions of the silanols with surface hydroxyl groups present on oxide surfaces [12], [15], [23], [24].

Hydrolysis:

H C(CH SiCl H O

H C(CH Si(OH) HCl

Condensation

Surface-OH+H C(CH Si(OH)

Surface-O-Si(OH) (CH CH H O Layers are generally ordered and closely packed, given sufficient surface hydroxyl group density [9], [10], [15]. Covalently bonded monolayer formation is generally claimed in the literature when the coverage (measured by FTIR spectroscopy) satu- rates with time following Langmuir kinetics, while an ordered layer results in a water contact angle of 112 [12], [13], [15], [16], [25].

C. Investigations Into the Kinetics of OTS Self Assembly Onto SF57

In order to assess irreversible OTS film formation on SF57 glass, we investigated the kinetics of the film formation on the polished 4 4 mm samples by directly dipping them into solution for the required time. This was followed by a cleaning stage, in which the samples were placed into warm 50 C analytical grade toluene ( 99% purity) in an ultrasonic bath for

20 min. The predominant reason for the inclusion of this stage is to remove any residual solvent molecules and unbound silane from the glass surface, ensuring that only bonded molecules re- main [26]. This also demonstrates the films robustness in a relatively harsh environment. The samples were analyzed using both FTIR spectroscopy and contact angle measurements as is frequently done in the literature for such SAMs [9]–[17], [25]–[27]. The FTIR spectra were measured in the attenuated total internal reflection mode, which is highly surface sensitive due to the short penetration depth of the evanescent field at the glass surface. We used the vibrational modes of the hydro- carbon tail (the CH stretches in particular) for analysis of the film formation kinetics due to its strong signal compared with uncoated samples. Although the OH group is the functional end of both the glass and the silane, it is not useful for quantitative analysis. Aside from producing a weaker signal than the CH band, bonded silane will at best remove one OH group from the surface, sometimes introducing up to two additional groups de- pending on the degree of cross-linkage in the film. There will

(3)

Fig. 3. Sealing of fiber to pump (the pump is attached to the needle), and the filling rate measurement setup.

therefore be no net change on average, in the number of OH groups present within measurement uncertainty.

To provide further insight in the film formation, we measured contact angles using a small stage to mount and position the sample below a microliter syringe, in front of a CCD camera. A small drop of water is placed on the surface, and a picture taken, which is used to determine the contact angle. This is repeated no less than 5 times for each sample to ensure reproducibility.

D. Filling, Flushing and Emptying the Fiber

As MOFs contain micron sized holes these act as small capil- laries with relatively large capillary forces when dipped into solution. MOFs will fill naturally due to these forces, and in particular, if the MOF is horizontal any length of fiber can in principle be filled. We have developed a novel method for measuring the fill rate of MOFs, and have the ability to apply a pressure difference across the length of the fiber holes. As will be shown, this is advantageous as MOFs fill relatively slowly compared with the deposition kinetics of OTS.

The filling rate is measured by a novel method recently developed within our research group [28]. Briefly, a low concentration (1 mM or less) rhodamine-B dye solution is allowed to fill the fiber via capillary action from one end. 532 nm (green) light from a diode laser is launched into the fiber core which then ex- cites the dye present in the holes. The rhodamine-B fluoresces red, and at the liquid air interface, a bright yellow spot is seen to move up the fiber. Alternatively a green optical filter can be used to locate the red spot. Unfilled fiber glows bright green due to scattering and loss through the cladding. Filled fibers appear almost entirely dark due to absorption by the dye. This allows the fill rate to readily be measured with an uncertainty in the spot position of approximately 5 mm. The setup is schematically represented in Fig. 3. The model of capillary filling by Wash- burn [29] is used to compare with experimental data. Washburn begins with Poiseuille’s law for a circular tube

(1) where is the volume of fluid in the capillary, its radius, and the fluid viscosity. is the capillary length filled (i.e., fiber length filled) at time , the coefficient of slip, and is the total driving pressure (i.e., the sum of all pressure sources acting on the fluid). Including pressures for a horizontal tube at a depth in the beaker and integrating with respect to time yields

the following result for the time dependence of the fiber length filled:

(2) where , , and are the liquid density, surface tension and contact angle, is acceleration due to gravity, and is any external pressure difference between the capillary ends.

E. Fiber Coating

Three stages were employed to coat the MOF, and each stage is performed using a pressure difference to force fluids through the fiber at an increased rate. The fiber is sealed at one end using a rubber stopper (suba-seal) and glass tube, which is connected to a small diaphragm pump via a needle (see Fig. 3). This end will be referred to as the “pump end” throughout the paper. The other end of the fiber (referred to as the “solution end”) is placed into a beaker containing the required deposition solution. The exact pressure difference introduced is not measurable with the current setup; however, the diaphragm pump is capable of producing a vacuum of 1300 Pa.

The first stage is self assembly of the OTS film in which the silane solution is passed through the fiber and sufficient time is given to allow the SAM to form. The pump is then stopped and the cleaning (second) stage begins by replacing the silane solution with toluene and restarting the pump. Following this, the fiber is removed from the liquid and air is pumped through, drying the surface of the holes (the third stage). A short piece of fiber ( 50 cm) was coated initially using the following conditions, chosen based on the results for polished 4 4 mm samples: 90 min on stage one, 30 min on stage two, and 12 min on stage three. The fiber was then carefully removed from the rubber seal, and a second, longer piece of fiber ( 1.2 m) coated with the following conditions: 150 min (stage one), 60 min (stage two), and 30 min (stage three). Approximately 5 cm was cleaved from both ends of each piece of fiber to ensure an uncoated cleaved end-surface is present.

III. RESULTS ANDDISCUSSION

A. FTIR and Contact Angle Results

As mentioned, we focus on the CH stretch absorption band in the FTIR spectra, as a relative measure of the surface coverage

(4)

Fig. 4. Example FTIR-ATR spectra of glass samples coated for different deposition times with OTS. The band of peaks represents the alkane stretches. For analysis we use the area under the CH2 Asymmetric stretch.

Fig. 5. Kinetics of the deposition of OTS onto lead silicate glass. The solid line represents a Langmuir kinetic response, and is provided for comparison.

of OTS. Due to the long alkane chain, this is the strongest absorption feature of the OTS molecule, and we give an example for a series of 4 4 mm samples in Fig. 4. The spectrum of an uncoated sample has been subtracted from all traces in the figure giving the relative increase in absorption for different deposition times.

The area under the CH asymmetric stretch peak of the FTIR spectra versus dipping time is shown in Fig. 5. The error bars represent the experimental uncertainty associated with the measurement technique; that is a sum of both the instrument uncertainty and the uncertainty associated with repeated measurements of a given sample. The data point is an average of these measurements. Basic adsorption theory predicts Langmuir kinetics ( where is the coverage, is a rate constant and is time) for irreversible film adsorption to a substrate [9]–[17], [25]–[27]. The solid line in Fig. 5 represents a

Fig. 6. Water contact angle on OTS coated lead silicate glass for different deposition times. The dashed line (90 ) represents the cross over between a wetting and nonwetting surface.

Langmuir kinetic response. Although oscillatory adsorption of silanes has been shown to occur [30]–[32], it has been not been observed for OTS due to its relatively large molecular mass [31].

Five deposition time data points (20, 60, 100, 140, and 180 min) were produced in triplicate to investigate reproducibility of the film formation. It is clear that there is a significant variance in the triplicate measurements, and given this, the data lies within the Langmuir regime. The observation of Langmuir adsorption kinetics demonstrates the irreversible chemisorption of OTS via covalent bonding to our SF57 glass samples, as reported in the literature for silica substrates [9]–[17], [25]–[27]. Contact angles for selected samples are also given in Fig. 6. The dashed line at 90 represents the crossover between a wetting and nonwetting surface, which for water implies hydrophilic and hydrophobic surfaces respectively. One 60 min sample shows a

(5)

contact angle of 112 indicating a closely packed layer. How- ever, in the majority of cases, the contact angle is lower than that observed for silica indicating a lower degree of packing and/or coverage. For the purpose of this study, the critical feature is that a hydrophobic surface is readily obtained on SF57. This implies that the (relatively high) surface tension of the glass is drasti- cally reduced to below that of water, resulting in nonwetting behavior. Studies of OTS films on silica have shown the surface tension of the film to be 20–30 mN/m [15], whilst water has a surface tension of approximately 70 mN/m [33] at the same temperature.

B. Capillary Flow Rate for Uncoated Fiber

We have measured the capillary fill rate of both isopropyl alcohol (IPA) and deionized water for an uncoated SF57 suspended core MOF using the method outline in Section II. The experimental data are given in Fig. 7(a) while Fig. 7(b) plots versus showing excellent agreement with the model in [29].

We performed a fit of the data in (a) and extracted the coefficient of slip using the physical properties of both liquids taken from [33] at 20 C, taking the contact angle as that measured on the flat samples (IPA contact angle 5 ). The radius of the capillary is approximated by taking the area of one MOF hole, and equating it to the area of a circle which gives a 4 m radius. For water, the best fit is achieved for a slip coefficient of , while for IPA the coefficient seems negli- gible . This can be qualitatively understood through the following reasoning. The slip coefficient can be defined as the ratio of viscosity to the frictional force coefficient [34]. The larger surface tension (and therefore contact angle) of water implies a lower friction coefficient compared to IPA, while their viscosities are of the same order. We thus expect water to exhibit a larger slip.

C. Capillary Flow Rate for Coated Fibers

To verify the existence of a thin film within the MOF holes, consider the extreme case of water and a complete closely packed layer. The glass surface will be nonwettable and a capillary depression will occur akin to mercury in a glass tube, i.e., no filling will be seen. For less complete layers with higher surface tension(s), the filling rate should be dramatically reduced.

Thus, repeating the fill rate measurement for the coated pieces will verify the presence of the film, and constitutes a positive test for functionalization of the surface. The shorter piece of fiber was tested first by placing it into a water/dye solution as described in Section II. Both ends were tested and after more than 12 hours in the solution there was still no visible fluorescence and no dimming of scattered light due to absorption by the dye from either end. In the unlikely case that the presence of the rhodamine dye significantly altered the water surface tension, therefore this result was visually confirmed under an optical microscope where it is possible to clearly see a (pure) water droplet drawn up by an uncoated fiber. IPA, which has a comparable surface tension ( 22 mN/m) [33] to the OTS films [15] was seen to enter this short piece of fiber under the microscope. However, due to breakage this piece was too short at this stage to conduct a quantitative fill rate measurement.

Fig. 7. (a) Filling rate of Water and IPA for coated and uncoated MOF. The lines are theoretical fits for IPA and water calculated using the liquid properties at standard temperature and pressure. (b) Plots ofl versustfor uncoated and coated fiber. The lines are linear fits with the correspondingR value, con- firming the model presented by Washburn [31].

The second, longer piece of fiber was then tested with no filling of water observed from the solution end of the fiber.

The pump end, however, showed dramatically reduced capillary filling for water. This was measured, along with IPA filling rates, and the results are given in Fig. 7(a). A dramatically lowered filling rate is observed for both water and IPA, indicating a significant modification to the MOF surface, and a positive test for film presence. Whilst IPA is expected to fill the fiber, since it has a surface tension comparable to that of the film, it is clear that the pump end of the fiber has an incomplete layer, with the surface still wettable by water. Although the contact angle is unknown for this incomplete layer, and thus becomes a free parameter for a theoretical fit, Fig. 7(b) confirms the dependence for the coated fiber allowing us to explore its possible values; and we expect the water contact angle to be close to the 90 limit (any higher and a capillary depression would occur) for the following reason. While we can no longer extract the coefficient of slip, the smallest contact angle possible would occur for zero slip, and for this case a fit [shown in Fig. 7(a)] gives 89.8 . Any nonzero slip will only increase this towards the 90 limit, and thus the possible range of contact angles agrees with our observation. Notice that the filling rate of IPA is lower

(6)

than water in all cases as expected due to its lower surface tension, and in particular, is lower from the solution end of the fiber supporting the claim of a less complete (higher surface tension) layer at the pump end of the fiber.

Although the effect of a thickness gradient in the film was not observed with the shorter piece of fiber, it is expected to occur for the following reasons. According to Langmuir kinetics, OTS should ideally form only a monolayer, and the glass surface exposure time past saturation should not be of concern. In prac- tice this is not so. Hydrolyzed silanes are able to cross link to one another, via the same condensations reactions bonding them to the fiber surface. As dimers and larger units form [26], a cloudy solution results due to polymerization lowering its concentration and reducing the deposition rate. Dimers (and larger units still in solution) also facilitate a less ordered film formation (with lower contact angle(s)) as well as multilayers [26], [27], with these effects strongly dependent on moisture content in the atmosphere and in solution [15], [23]. Indeed, for our ex- periment, the solution displayed cloudiness towards the end of the 150 min solution stage of deposition. One can avoid this by working under a controlled atmosphere; however, this was not done in this study. The lowered concentration coupled with a reduced exposure time at the pump end results in an effective deposition time well below saturation.

Thus the observed thickness gradient in the film is very rea- sonable, with a more complete, possibly multilayer film forming at the solution end, and an incomplete, less ordered layer toward the pump end. Without the introduced pressure difference, which significantly increases the fill rate, we predict this effect would be much more significant and may have been observable in the shorter piece. Given the success we have observed in ap- plying the model by Washburn, one could use this model to gain an estimate of the increase in filling rate as a result of the external pressure difference by simply including it in (2) (the coefficient of slip would also be affected however). It should be noted that even though this effect was not observed in the shorter piece, a thickness gradient is still extremely likely. The impli- cation is simply that the surface is not wettable from either end.

IV. CONCLUSION

We have investigated the adsorption properties of self assembled monolayers of octadecyltrichlorosilane onto lead silicate glass and found similar behavior as that seen for silica. Based on these results we have successfully demonstrated that the internal surfaces of a lead silicate microstructured optical fiber can be functionalized with such self assembled films of silane coupling agents. The capillary fill rate of our fibers is measured for both water and isopropyl alcohol via a novel method recently developed at our centre, and we measure fill rates which agree strongly with theory. It is shown that a comparison of fill rate before and after application of the silane can be used as a positive test for surface functionalization. The functionalized fibers have a dramatic reduction in measured fill rates, with the extreme case of no capillary filling (i.e., a capillary depression) for nonwettable layers observed, as a result of the silane films’ low surface tension. This work provides the first step toward further research into more exotic functionalization of microstructured

fibers in applications such as biophotonics and chemical fiber sensors, as well as increased glass protection from water degradation. Although the method has been presented for a specific fiber geometry and glass, it could be translated with ease to other geometries and (oxide) glasses.

ACKNOWLEDGMENT

The authors wish to thank S. C. Warren-Smith for his help with the uncoated fiber fill rate measurements, and R. C. Moore for making the fiber.

REFERENCES

[1] J. C. Knight, T. A. Birks, P. St. J. Russell, and D. M. Atkin, “All-silica single-mode optical fiber with photonic crystal cladding,”Opt. Lett., vol. 21, no. 19, pp. 1547–1549, Oct. 1996.

[2] T. M. Monro and H. Ebendorff-Heidepriem, “Progress in microstructured optical fibers,”Ann. Rev. Mater. Sci., vol. 36, pp. 467–495, Aug.

2006.

[3] J. B. Jensen, L. H. Pedersen, P. E. Hoiby, L. B. Nielsen, T. P. Hansen, J. R. Kolkenberg, J. Riishede, D. Noordegraaf, K. Nielsen, A. Carlsen, and A. Bjarklev, “Photonic crystal fiber based evanescent-wave sensor for detection of biomolecules in aqueous solutions,”Opt. Lett., vol. 29, no. 17, pp. 1974–1976, Sep. 2004.

[4] J. B. Jensen, P. E. Hoiby, G. Emiliyanov, O. Bang, L. H. Pedersen, and A. Bjarklev, “Selective detection of antibodies in microstructured polymer optical fibers,”Opt. Exp., vol. 13, no. 19, pp. 5883–5889, Jul.

2005.

[5] T. M. Monro, W. Belardi, K. Furusawa, J. C. Baggett, N. G. R. Brod- erick, and D. J. Richardson, “Sensing with microstructured optical fibers,”Meas. Sci. Technol., vol. 12, no. 7, pp. 854–858, Jul. 2001.

[6] A. K. Varshneya, Fundamentals of Inorganic Glasses. San Diego, CA: Academic Press, 1994.

[7] D. C. Tran, G. H. Sigel, Jr., and B. Bendow, “Heavy metal fluoride glasses and fibers: A review,”J. Lightw. Tecnol., vol. 2, no. 5, pp.

566–586, Oct. 1984.

[8] A. F. Kosolapov, S. L. Semjonov, and A. N. Denisov, “Mechanical properties of high-purity silica microstructured optical fiber,”Inorg.

Mater., vol. 43, no. 3, pp. 310–314, Mar. 2007.

[9] J. D. Le Grange and J. L. Markham, “Effects of surface hydration on the deposition of silane monolayers on silica,”Langmuir, vol. 9, no. 7, pp. 1749–1753, Jul. 1993.

[10] J. Gun, R. Iscovici, and J. Sagiv, “On the formation of self-assembling monolayers (II),”J. Colloid Interfac. Sci., vol. 101, no. 1, pp. 201–213, Sep. 1984.

[11] J. Gun and J. Sagiv, “On the formation of self-assembling monolayers (III),”J. Colloid Interfac. Sci., vol. 112, no. 2, pp. 457–472, Aug. 1986.

[12] R. Maoz and J. Sagiv, “On the formation of self-assembling monolayers (I),”J. Colloid Interfac. Sci., vol. 100, no. 2, pp. 465–496, Aug.

1984.

[13] S. R. Wasserman, G. M. Whitesides, I. M. Tidswell, B. M. Ocko, P. S.

Pershan, and J. D. Axe, “The structure of self-assembled monolayers of alkylsiloxanes on silicon: A comparison of results from ellipsometry and low-angle x-ray reflectivity,”J. Amer. Chem. Soc., vol. 111, no. 15, pp. 5852–5861, Jul. 1989.

[14] S. S. Cheng, D. A. Scherson, and C. N. Sukenik, “In situ observation of monolayer self-assembly by FTIR-ATR,”J. Amer. Chem. Soc., vol.

114, no. 13, pp. 5436–5437, Jun. 1992.

[15] D. L. Angst and G. W. Simmons, “Moisture absorption characteristics of organosiloxane self-assembled monolayers,”Langmuir, vol. 7, no.

10, pp. 2236–2242, Oct. 1991.

[16] R. Banga, J. Yarwood, A. M. Morgan, B. Evans, and J. Kells, “FTIR and AFM studies of the kinetics and self-assembly of alkyltrichlorosilanes and (perofluroalkyl)trichlorosilanes onto glass and silicon,”Langmuir, vol. 11, no. 11, pp. 4393–4399, Nov. 1995.

[17] R. Banga, J. Yarwood, and A. M. Morgan, “Determination of the or- dering of alkylhalogensilanes on a silicon substrate using FTIR-ATR spectroscopy,”Langmuir, vol. 11, no. 2, pp. 618–622, Feb. 1995.

[18] T. M. Monro, Y. Ruan, H. Ebendorff-Heidepriem, T.-C. Foo, P. Hoff- mann, and R. C. Moore, “Antibody immobilization within glass microstructured fibers: A route to sensitive and selective biosensors,” presented at the Opt. Fiber Sens. Conf. (OFS’2008), Perth, Australia, Apr.

2008.

(7)

[19] S. C. Warren-Smith, S. Afshar, and T. M. Monro, “Theoretical study of side access and thin-layer fluorescence-based optical-fibre sensors with sub-wavelength features,” presented at the Opt. Fiber Sens. Conf.

(OFS’2008), Perth, Australia, Apr. 2008.

[20] H. Ebendorff-Heidepriem and T. M. Monro, “Extrusion of complex preforms for microstructured optical fibers,”Opt. Exp., vol. 15, no. 23, pp. 15086–15092, Nov. 2007.

[21] J. Y. Y. Leong, P. Petropoulos, J. H. V. Price, H. Ebendorff-Heide- priem, S. Asimakis, R. C. Moore, V. Finazzi, K. E. Frampton, X. Feng, T. M. Monro, and D. J. Richardson, “High-nonlinearity dispersion- shifted lead-silicate holey fibers for efficient 1-m pumped supercon- tinuum generation,”J. Lightw. Technol., vol. 24, no. 1, pp. 183–190, Jan. 2006.

[22] H. Ebendorff-Heidepriem, P. Petropoulos, V. Finazzi, S. Asimakis, J.

Leong, F. Koizumi, K. Frampton, R. C. Moore, D. J. Richardson, and T. M. Monro, “Heavy metal oxide glass holey fibers with high nonlinearity,” presented at the Opt. Fiber Sens. Conf. (OFS’2008), Anaheim, CA, Mar. 2005, paper OThA3.

[23] P. Silberzan, L. Léger, D. Ausserré, and J. J. Benattar, “Silanation of silica surfaces. A new method of constructing pure or mixed monolayers,”Langmuir, vol. 7, no. 8, pp. 1647–1651, Aug. 1991.

[24] E. P. Plueddemann, Silane Coupling Agents, 2nd ed. New York:

Plenum Press, 1991.

[25] K. M. R. Kallury, M. Thompson, C. P. Tripp, and M. L. Hair, “In- teraction of silicon surfaces silanized with octadecylchlorosilane with octadecanoic acid and octadecanamine studied by ellipsometry, x-ray photoelectron spectroscopy, and reflectance Fourier transform infrared spectroscopy,”Langmuir, vol. 8, no. 3, pp. 947–954, Mar. 1992.

[26] A. Baptiste, A. Gibaud, J. F. Bardeau, K. Wen, R. Maoz, J. Sagiv, and B. M. Ocko, “X-ray, micro-Raman, and infrared spectroscopy struc- tural characterization of self-assembled multilayer silane films with variable numbers of stacked layers,”Langmuir, vol. 18, no. 10, pp.

3916–3922, May. 2002.

[27] R. Maoz, J. Sagiv, D. Degenhardt, H. Möhwald, and P. Quint, “Hy- drogen bonded multilayers of self-assembling silanes: Structure elu- cidation by combined FTIR spectroscopy and x-ray scattering tech- niques,”Supramol. Sci., vol. 2, no. 1, pp. 9–24, Mar. 1995.

[28] S. Afshar, V. S. C. Warren-Smith, and T. M. Monro, “Enhancement of fluorescence-based sensing using microstructured optical fibres,”Opt.

Exp., vol. 15, no. 26, pp. 17891–17901, Dec. 2007.

[29] E. W. Washburn, “The dynamics of capillary flow,”Phys. Rev., vol. 17, no. 3, pp. 273–283, Mar. 1921.

[30] J. S. Quinton, “The oscillatory adsorption of organofunctional silanes on metal oxide surfaces,” Ph.D. dissertation, Univ. Newcastle, New- castle, U.K., 2000.

[31] L. Houssiau and P. Bertrand, “ToF-SIMS study of organosilane self- assembly on aluminium surfaces,”Appl. Surf. Sci., vol. 175–176, pp.

351–356, May. 2001.

[32] L. Houssiau and P. Bertrand, “ToF-SIMS study of organosilane adsorption on model hydroxyl terminated surfaces,”Appl. Surf. Sci., vol.

203–204, pp. 580–585, Jan. 2003.

[33] J. A. Dean, Lange’s Handbook of Chemistry, 10th ed. New York: Mc- Graw-Hill, 1967.

[34] L. B. Loeb, The Kinetic Theory of Gases. New York: Dover, 2004, pp. 281–285.

John E. Debsreceived the B.Sc. degree (first class hons) in physics and nanotechnology and a Uni- versity medal from Flinders University, Adelaide, Australia, in 2006. His honours year was conducted working closely in collaboration with the Centre of Expertise in Photonics at the University of Adelaide, Australia, pioneering work on surface functionalization of soft glass microstructured optical fibers. He is currently working towards the Ph.D. degree with the Australian Centre of Excellence for Quantum Atom Optics within the Department of Physics, Australian National University, Canberra, Australia.

His current research interests include laser systems for atoms optics, tunable interactions in Bose–Einstein condensates, and applications of the atom laser.

Heike Ebendorff-Heidepriemreceived the Ph.D.

degree in 1994 from the University of Jena, Jena, Germany, on spectroscopy of rare-earth-doped glasses.

From 1994 to 2000, she was with the Otto Schott Institute for Glass Chemistry, University of Jena, Germany, working on new optical glasses. From 2001 to 2004, she was with the Optoelectronics Research Centre, University of Southampton, U.K., working on novel photosensitive glasses and soft glass microstructured optical fibers with record high nonlinearity. She has been leading the glass science and fabrication technology research within the Centre of Expertise in Photonics, School of Chemistry and Physics, University of Adelaide, Australia, since 2005. Her current research focuses on the development of new soft glasses and microstructured optical fibres for sensing, high power generation and transmission, and nonlinearity applications.

Jamie S. Quintonreceived the Ph.D. degree on the surface modification of metal oxides with organofunctional silanes for corrosion protection from the University of Newcastle, Newcastle, Aus- tralia.

From 2000 to 2001, he worked as a Postdoctoral Fellow with Prof. T. Madey at the Laboratory for Surface Modification, Rutgers University, New Brunswick, NJ, where his role was to lead the group’s synchrotron research. He is a Senior Lec- turer in Physics and Nanotechnology at Flinders University, Adelaide, Australia and is a co-founder of the Smart Surface Structures Research Group. His research interests cover many applications of surface modification, from carbon surfaces for molecular electronic devices and sensors, through to optically active coatings with dyes and quantum dots.

Tanya M. Monroreceived the Ph.D. degree on self-written waveguides in photosensitive glasses from the University of Sydney, Sydney, Australia.

From 1998 to 2004, she was with the ORC, Univer- sity of Southampton, Southampton, U.K., working on silica and soft glass microstructured optical fibers.

She has been the Chair of Photonics and the Director of the Centre of Expertise in Photonics within the School of Chemistry and Physics, University of Ade- laide, Adelaide, Australia, since 2005. Her current research within the Centre of Expertise in Photonics focuses on the design, fabrication and device applications of new classes of soft glass microstructured optical fibers, including work on new transmission fibers, highly nonlinear fibers, chem/bio sensing with new fibers and novel fiber lasers.