Open Access
Add to BibSonomyAbstract
A new ethanol vapor detection probe based on an optical fiber long period grating overlaid with a zinc oxide (ZnO) nanorods layer is presented. The ZnO nanorod layer was developed onto the cladding of the fiber using aqueous chemical growth, seeded by a thin layer of metallic Zn. The growth of the ZnO nanorods overlayer onto the long period grating cladding is monitored in real time for investigating its effect on the spectral properties of the device and its subsequent role in the sensing mechanism. Results are presented, on the correlation between the growth time of the ZnO layer and the ethanol vapor detection performance. Reversible spectral changes of the notch extinction ratio of more than 4dB were recorded for ~50Torr of ethanol vapor concentration. In addition, photoluminescence emission studies of the ZnO overlayer performed simultaneously with the optical fiber spectral measurements, revealed significant ethanol induced changes in the intensity of the bandgap peak.
©2012 Optical Society of America
1. Introduction
Detection and monitoring of gaseous or volatile chemicals, employing optical fiber sensors, has received considerable attention in recent years, thanks to a number of features such as compactness, high sensitivity and remote operation, easily achieved with this technology, that open possibilities for demanding analytical and process monitoring applications. Such sensing probes are now used in diverse applications, for example, toxic and warfare gas detection [1] or environmental monitoring [2], when accurate characterization and long-term sampling is required to reduce health risks and/or ensure public safety. Different optical fiber interrogation schemes and sensing mechanisms have been proposed and implemented, targeting specific requirements of each particular application. One of the most efficient sensor architectures employs optical fiber long-period gratings (LPGs) overlaid with thin coatings of materials having specifically tailored properties that enable detection of a number of different chemical species at a wide range of concentrations [3–7]. More specifically, LPGs operate by coupling light from the core mode of a single mode fiber into cladding modes and this is manifested as attenuation bands observed at specific wavelengths in the transmission of the fiber. Because the evanescent wave of the cladding modes exceeds the cladding area, LPGs are sensitive to the optical properties of materials surrounding the fiber. This distinctive feature has led to extensive studies of LPGs in conjunction with out-cladding coatings, engineered to change their optical properties in response to an external stimulus.
In this context, metal oxides are known to undergo variations of their optical and/or electronic properties upon chemisorption, rendering themselves ideal for various chemical sensor applications and sensing schemes [8–10]. Zinc oxide (ZnO) is one of the most promising materials for the development of chemical probes [11–13] not only because of its favourable optical, electrical and mechanical properties but also because of the diversity of synthesis methods available and the variety of topologies that can be grown (thin films, nano-rods, nano-particles etc). ZnO is a wide band gap n-type semiconductor material that exhibits chemical and thermal stability, high transparency in the visible and near infrared, and strong photoluminescence emission that enables lasing [14] In addition, ZnO can be doped with a number of metals that may influence optical, electrical and magnetic properties [15]. Regarding its chemosensitive performance, ZnO has been successfully used in the detection of ozone, gasoline, toluene and ethanol [9–11] on the basis of electrical interrogation exploiting the significant conductivity changes induced on its surface upon chemisoprtion of an organic vapor or active gas. Recently, ethanol sensing was utilized using ZnO thin films, interrogated by a surface Plasmon resonance scheme [16] but also ZnO nanoparticles overlaid onto micro-ring resonators [13]. Most of these ZnO chemical sensors have been proven capable of tracing volatile compounds within a wide dynamic range of concentrations, however, operating at elevated temperatures, typically greater than 150°C.
Focusing on the detection of alcohols, the straightforward and reliable determination of ethanol using miniaturized and low cost sensing probes is of great commercial importance since it refers to direct applications in the automotive and bio-fuel industry [17], forensics (detection of alcohol consumption in humans) [18] or wine and spirits industry [19]. Ethanol is used neat or mixed with gasoline as automotive fuel while at concentrations in the range of 4-16% v/v is found in wine and beer. In this context the development of ethanol sensing units, that need no heating elements and ensure broad dynamic range probing, preferably between 1 and 90% v/v vapor concentrations, can cover most of the aforementioned application fields.
Herein, a novel ethanol vapor detection probe is described, which employs ZnO nanorods as an out-cladding coating on a LPG. The probe operates at room temperature and is optically interrogated at the 1.5 μm band. In this disruptive approach, the use of ZnO nanorods, growing radially off of the fiber surface, instead of a film [13] offers enhanced gas sensing performance attributed to the higher surface-to-volume ratio exhibited by the nanorods as compared to that of coarse micrograined materials. The ZnO nanorods have been directly grown on the stripped optical fiber, following LPG inscription, through a two step process that involves i) deposition of a zinc precursor layer by pulsed laser deposition (PLD) and ii) aqueous chemical growth (ACG) of ZnO nanorods [10, 20]. When the ZnO coated LPG is exposed to ethanol vapor its optical properties, namely absorption loss and/or refractive index are altered accordingly. In fact, these alterations induced by the vapor on the ZnO nanorod layer are straightforwardly interrogated by the cladding mode of the LPG structure, and clearly manifested as strength and/or peak wavelength changes in its transmission spectrum.
In order to obtain an accurate observation of the ZnO nanorods growth onto the fiber cladding but principally for investigating the effect of the growth conditions on the spectral behavior of the LPG probe and its subsequent sensing behavior, the growth process was monitored on-line through the spectral modification induced on the Zn coated LPG [21, 22]. This monitoring step is of high importance and largely determines the performance of the LPG detection probe since the opto-geometrical characteristics (shape, length and spatial density), and physical properties (crystallinity and defect concentration) of the ZnO nanorods define the vapor detecting sensitivity. Through this procedure interesting observations related to the growth process are derived and most importantly LPGs with ZnO coatings of different characteristics are produced. Material characterization has been based manually on scanning electron microscopy (SEM, GEOL-JSM-7000F), x-ray diffractometry and photoluminescence measurements. SEM images of the overlaid optical fibers provide information related to the growth topology and size of the ZnO nanorods. Furthermore, photoluminescence studies of the overlaid ZnO nanorods carried out simultaneously with the LPG spectral measurements provide an insight to the electronic changes induced on the nanorods upon their exposure to ethanol. Through this manifold approach we attempt to investigate both the photonic and material parameters affecting the sensitivity performance of this ZnO based fiber optic ethanol sensor. Potentially, the photoluminescence results could be correlated with the optical fiber measurements towards the illustration of the underlying sensing mechanism [23]. While there are several studies and assertions related to the sensing mechanism of organic and inorganic gaseous substances using ZnO nanostructures operating at high temperatures, few of them address such a sensing event at room temperature [23, 24] as in the case presented here.
2. Experimental
For the realization of the detection probe LPGs were inscribed in a Boron co-doped germanosilicate fiber (PS1250/1500–Fibercore Ltd) utilizing a Spectra Physik KrF excimer laser, emitting pulses with 34ns duration at 248 nm [7]. A titanium foil amplitude mask, with a period of 407 μm was employed and the laser fluence incident on the fiber was 300 mJ/cm2 per pulse, at a repetition rate of 10 pulses/s. For an exposure time of approximately 100 seconds under the above conditions LPGs with up to 16dB strength and 20 mm length are inscribed with attenuation bands observable in the 1300-1700 nm region.
Growing ZnO nanorods onto the LPGs, is based on a two-step procedure [20] involving first pulsed laser deposition of a zinc (Zn) layer onto the optical fiber followed by aqueous chemical growth of ZnO nanostructures. For the deposition of the zinc precursor layer a 248nm KrF excimer laser beam was focused on a Zn target (99.9% purity) to yield a fluence of 1.3 J/cm2. The ablated material was collected on the optical fibers placed at a distance of 6 cm from the target surface. Depositions on flat substrates under identical conditions have resulted in Zn layers approximately 40 nm thick, after exposure with 2000 laser pulses. Spectral comparison of fiber transmission before and after Zn deposition showed that the Zn layer has a negligible effect on the LPG spectrum. Following deposition, the Zn seeded optical fibers were chemically treated in an aqueous environment, 0.05M solution of zinc nitrate hexahydrate (Zn(NO3)2(H2O)6) and ammonium hydroxide (NH3) [20]. More specifically, Zn coated optical fiber LPGs, mounted on a Teflon holder, were immersed in the ammoniac zinc hydrate solution at room temperature under constant stirring. The solution was gradually heated up to the optimum temperature of around 80°C which was then kept constant for up to 3 hours. After about 20 min time, and temperature of about 50οC the originally transparent and colorless solution assumes a “milky appearance”, signaling the onset of the ZnO nanocrystals growth, and remains so until the termination of the procedure.
The ZnO overlaid gratings were tested as for their ethanol sensitivity inside a specially designed stainless steel chamber of an internal volume of 500 ml, with appropriate inlets and outlets for injecting ethanol in the form of vapor or liquid and nitrogen buffer gas. Fused silica windows were fitted on the chamber for allowing photoluminescence measurements and visual inspection of the sample. The LPGs could be held under tension by a set of magnets positioned inside the chamber. The optical characterization measurements were performed using a superluminesent source and an Optical Spectrum Analyzer (OSA ANDO AQ6317B) for monitoring in real time the spectral response of the LPG. Each of the gratings was placed inside the chamber and a reference spectrum was recorded. Prior to ethanol injection, the chamber was flushed using nitrogen flow for eliminating residues of humidity that could affect the specificity and accuracy of the measurements. Following the introduction of ethanol, the LPG spectrum was monitored in constant interval of 2-5 min until saturation. Ethanol was introduced into the chamber either as a liquid (0.1-0.5 ml) that was left to evaporate or in the form of vapor, by use of bubbler; both cases produced comparable results. Using the Antoine equation [25] and for a temperature of 22°C measured in the chamber the saturated vapor pressure of ethanol was calculated to be ~50 Torr. Furthermore an ethanol resistivity measurement probe (PS-2194, Pasco) was used to monitor the relative change in the ethanol vapor concentration during measurements.
In parallel, and for gaining a deeper understanding of the effect of the ethanol vapor on the defect concentration of the ZnO overlayer, photoluminescence measurements were recorded. Accordingly coated LPGs in the chamber were excited at room temperature by the third harmonic, 355nm, of a 8 ns flash-lamp pumped, Q-switched Nd:YAG laser (Spectron Laser Systems, SL404), at energies between 0.1 and 1.9 mJ/pulse. The photoluminescence (PL) was imaged through the silica window of the chamber, onto a multimode silica glass optical fiber (at 1:1 magnification) by means of a UV fused silica lens system of f = + 25 mm. The fiber end was coupled into the entrance slit of a spectrograph (PTI, 01-002AD) equipped with a 300 lines/mm holographic grating. The resulting emission was recorded on an intensified charge coupled device detector (ICCD, Andor iStar, DH740-18F-03). Typically emission spectra were averaged over 150 laser shots. To study the dynamic behavior of the ZnO nanorods’ PL as a function of time upon exposure of the LPG to ethanol atmosphere, spectra were acquired at 5 min intervals and the integrated emission in a fixed wavelength range (370-500nm) was measured.
3. Real time monitoring of ZnO nanorod outcladding growth
Initial measurements on fibers withdrawn form the chemical growth solution at certain time intervals during the process showed that progressively and apparently because of nanorod formation the LPG attenuation bands (LPG notches) varied in both spectral peak position and strength. To obtain a more detailed picture of the growth process, the spectrum of the LPG was monitored in real time. For this reason light from a superluminescent diode was launched into the optical fiber while the output signal was recorded on the OSA. From the resulting spectra the change in the peak wavelength and strength of the attenuation bands was extracted. The results of the wavelength shift and the extinction ratio changes of the main LPG attenuation notch as a function of time are displayed in Fig. 1 . To determine the contribution of the temperature into the monitored LPG spectral changes, a pristine LPG (bearing no Zn layer) was also studied, under identical thermal and chemical conditions (see black circular-points in Fig. 1). The top graph in Fig. 1(a) depicts the temperature of the ACG solution versus time, measured using an immersion thermocouple. As shown in Fig. 1a the dominant perturbation behind the LPG peak wavelength blue shift occurring during the first 20min of the chemical growth, is temperature; this is clearly evident by the similar slope of the two curves representing the behavior of the Zn overlaid and the pristine grating. The visual change in the solution (increased scattering) taking place at 20min and denoting the formation of ZnO nanocrystals is predominantly accompanied by the onset of a further spectral blue-shift of the Zn overlaid LPG attenuation band. This blue shift is maintained up until to the 42nd min of the growth, then sharply reverses to a red-shift and soon after the 50th min, turns back to a blue shift that reaches a saturation plateau for prolonged time periods. Simultaneously, the notch strength of the Zn-overlaid LPG follows a gradually decreasing trend, due to the formation of a high refractive index outcladding layer, which retracts the cladding modal field closer to the glass-ZnO interface, lowering the overlap with the core mode [26]. The Zn overlaid LPG notch strength initially decreases, reaching its minimum value at ca. 45 min. Subsequently it gains strength without however retrieving its initial characteristics again. The corresponding LPG notch strength change of the pristine LPG registers only the onset of the ZnO growth process at 20 min and remains almost unaffected after that.
Fig. 1 (a) Wavelength shift of the LPG notch recorded during the chemical growth of ZnO. The x axis represent the time in the solution. (b) LPG notch strength change versus time in the solution.
Transmission spectra in the form of a contour plot corresponding to the effect of the growth process on the LPG behavior versus growth time in chemical solution are presented in Fig. 2 .
Fig. 2 Contour plot of the LPG transmission strength spectra versus growth time in the ZnO nanorods growth solution. Color bar represents grating intensity strength in dB.
As shown in Fig. 1 and 2 the growth of ZnO nanorods onto the fiber cladding, largely modifies the spectral response of the LPG, in both wavelength and strength, especially during the early stages of the growth process. In order to verify the existence of nanorods for such short growth time but also for gaining insight on the characteristics of the formed nanorods, ZnO coated LPGs were examined by scanning electron microscopy. As can be seen in Fig. 3 SEM images confirm the uniformity and long range continuity of the ZnO nanorod layer over the cladding area; see Fig. 3(a). The overlaid fibre cross-section in Fig. 3(b) shows no dangling nanorods or other debris deposition. Additional material oriented investigations included XRD measurements at θ/2θ configuration for checking the crystallinity of the ZnO grown nanorods versus growth time. All the overlayed LPGs exhibited the θ/2θ XRD lines that can be indexed to the hexagonal würtzite structure of ZnO, however because of the small size of the fiber and the short growth time [8] the intensity of the collected signal is rather low, rendering comparison of different overlayers inconclusive.
Fig. 3 SEM images of ZnO nanorod layers grown onto the LPG cladding under different views and under different chemical growth durations. (a) General view of ZnO narorod overlaid grating after 3-hour immersion in the ACG solution. The fibre has been intentionally scribed for illustrating the ZnO overlayer thickness. (b) Cross-section of a ZnO nanorods overlaid LPG; growth conditions as in (a). (c) Close view of ZnO overlayer after 27 min and (d) 58 min of chemical growth.
It is clearly concluded on the basis of the real-time monitoring of the ZnO growth process, that growth time strongly affects the spectral characteristics of the overlaid LPGs. Since this spectral signature will determine the sensing behavior of the resulting probes it was decided that overlaid LPGs of different growth times should be produced and examined. To facilitate this study, five LPGs bearing identically prepared Zn precursor layers, were positioned in the growth solution in a holder that allowed selective removal of each LPG during the growth process. The LPGs were removed from the solution at predetermined times, namely 27, 45, 58, 65 and 100 min. The ZnO coated LPGs were then washed using de-ionised water to remove the excess ZnO and the overlaid LPGs were left to dry in open air. SEM images for LPGs immersed into the solution for 27min and 58min are shown in Fig. 3(c), (d). and reveal the existence of well-developed nanorod structures even for growth times as short as 27 min and temperatures as low as 67°C. The short heighted (~100nm) and densely networked nanorods grown for 27 min immersion in the growth solution seem to undergo a quadratic increase in height when immersion time is doubled, while exhibiting a less packed density closer to the nanorod peaks.
4. Ethanol probing
Following ZnO chemical growth, the nanorod overlaid LPGs were tested for ethanol vapor detection. Initial studies were aimed at correlating efficiency of ethanol vapor detection with growth time and hence quality and thickness of the ZnO nanorod layer. In this respect overlaid LPGs corresponding to different growth times were tested by exposure to a controlled ethanol vapor atmosphere.
4.1 Detection of ethanol using the ZnO nanorod overlaid LPG
Initially the ZnO coated LPG immersed into the chemical growth solution for 58 min was tested and the change in the LPG notch strength induced in the presence of up to 50 Torr of ethanol vapor is presented in Fig. 4(a), (b) . As shown, in the presence of ethanol vapor the strength of the LPG attenuation band is decreasing, while the maximum induced LPG strength change is ~4.4 dB, under a saturated ethanol vapor pressure of 50 Torr; see Fig. 4(c). Correspondingly, a 2 nm blue shift in the peak wavelength of the transmission notch was measured during the same exposure; see Fig. 4(b). By comparing the response trend of the commercial ethanol probe, and the changes measured in the grating strength, one can see that the ZnO based sensor exhibits similar temporal response. Grating strength changes of the order of ~0.5 dB are easily measurable, thus, allowing the possibility of detecting ethanol vapors at lower concentrations.
Fig. 4 (a) Changes in the LPG notch strength ΔS and (b) wavelength upon exposure to 50Torr ethanol vapor versus exposure time. The solid line in graph (a) corresponds to the relative ethanol concentration as that was measured by the commercial PASCO probe. (c) Spectral effect of the ethanol vapor on the LPG attenuation band after 95min of exposure and corresponding recovery behavior to air.
Once spectral changes of the LPG notch saturated, the chamber was opened, exposing the sample to air, and within a typical time of 20s the LPG fully recovered back to its initial spectral strength; see green line data in Fig. 4(b). In contrast, the wavelength shift was permanent and the notch maximum did not shift back to the initial position after re-exposure to air. We speculate that this permanent wavelength shift may be associated with passivation of surface and volume defects existing into the ZnO nanorods by OH- species, induced by ethanol adsorption and assisted by the low operation temperature of the experiment [27]. The ‘air-ethanol-air’ cycle was repeated for a number of times and the LPG notch strength was modified according to the ethanol vapor concentration in the chamber as illustrated in Fig. 5 without any further effect on the wavelength of the LPG. The results of Fig. 5 confirm the repeatability of operation of the ZnO sensing probe for concatenated cycles, when the strength of the LPG notch is interrogated.
Identical investigation procedure was followed in the studies of ZnO overlaid LPGs corresponding to different growth times and the results in terms of ethanol vapor sensing are illustrated in Table 1 . Ethanol detection performance is compared on the basis of the induced change recorded in the LPG notch strength since for all reported samples, except the one corresponding to 58 min of nanorod growth mentioned above, negligible LPG wavelength shift was measured. For the LPG with the minimum growth time (27 min) the ethanol vapor detection efficiency is negligible while it gets stronger as the growth time increases and reaches a maximum value for the 58 min sample. For longer immersion time, the saturated LPG notch strength change, start to decrease again and the probe becomes insensitive for prolonged growth time. It is evident that there is an optimum growth time for maximum ethanol detection sensitivity and that the time window leading to optimum sensing performance is rather quite narrow. The growth time of the ZnO nanorod overlayer dominates the capability of the LPG to interrogate the ethanol induced optical changes into the nano-crystalline material. Simultaneous photoluminescence measurements (see also next section) performed in ZnO samples fabricated using immersion times from 20min to 100min into the chemical solution, shown that all the ZnO nanorods exhibit sensitivity to ethanol stimulus irrespectively fabrication time.
Table 1. LPG notch strength change in 50 Torr of ethanol atmosphere for different ZnO growth times
To explain the performance of such an overlayed LPG structure, generally a thin film overlayer model can be used [26], in order to illustrate how the overlayer influences the cladding modal fields and thus the scattering and sensing behavior of the LPG. However in the case of the semi-random oxide structure employed here this is not possible, since the spatial density, height and refractive index of the ZnO nanorod outcladding cannot be estimated accurately. Instead, a different approach was adopted to simulate the effect of the ZnO overlayer and assist in the interpretation of the results. Specifically the wavelength shift of an identical LPG was calibrated by using different refractive index liquid outcladdings and the results are presented in Fig. 6 for refractive index values in the range of 1.0 - 1.7. From this well documented response [28] the wavelength shifts for operation in the optimum steep slope region just below the refractive index of silica glass can be identified. Following the dynamic monitoring of the ZnO growth we have determined the wavelength shift induced on the LPG due to different ZnO overlayers, combining this information and the calibration of Fig. 5 we can justify the sensing performance of the resulting detection probes. These correlation measurements serve as a practical guideline for understanding the ethanol sensing behavior for LPGs overlaid with ZnO nanorods layers of different thickness.
Fig. 6 Change in the LPG notch wavelength due to outcladding liquids of different refractive index (RI) values. Blue cell points are corresponding oil outcladding refractive index measurement shifts measured for gratings developed for different times in the chemical solution.
Table 2 summarizes the growth conditions of the five ZnO overlayed LPG samples produced. Along with the growth time and temperature of each sample the induced “dry” wavelength shifts are given. “Dry” refers to the wavelength shift measured after the sample is withdrawn from the growth solution, rinsed and dried and illustrates the sole effect of the different ZnO overlayers on the LPG in contrast to the “wet” shifts (shown in Fig. 1) that are due to the combined effect of the overlayer and the chemical solution the LPG is immersed in. Employing the “dry” wavelength shifts and the experimentally determined relation between induced shift and overlayer refractive index (Fig. 6) a corresponding oil outclading refractive index can be deduced as illustrated in Table 2.
Table 2. Notch wavelength shift variations of ZnO nanorod overlaid LPGs in different phases of the growth process.
As can be seen from the data listed in Table 2, the LPG immersed in the chemical solution for 58min, exhibits the greatest “dry” wavelength shift, and corresponds to an oil outclading refractive index of approximately 1.402. This value is within the region of optimum outcladding refractive index values just below that of silica glass as illustrated by the steep slope in the graph of Fig. 6. It is therefore justified that this probe should exhibit increased sensitivities with respect to the ethanol stimulus as shown in the data presented in Table 1.
4.2 Photoluminescence (PL) measurements
The dynamic behavior of the sensors was probed also through photoluminescence emission measurements performed on the ZnO coated LPGs simultaneously with the optical interrogation during exposure of the fibers to ethanol atmosphere. Fibers were irradiated in the chamber at 355 nm by nanoseconds laser pulses giving rise to a strong UV photoluminescence emission band, centered at 390 nm, (Fig. 7 ) which is the characteristic near-band edge transition of the wide band gap of the semiconductor [29]. Introduction of ethanol into the chamber results in a distinct increase of PL intensity (red line) compared to that recorded on atmospheric air, in accordance with observations reported in [23]. It is noted that no peak wavelength shifts or band profile changes were observed. The increase of the spectral intensity but not of the corresponding spectral shape or maximum of the PL, possibly denotes that the presence of ethanol does not affect the type of defects excited, but primarily the defect population.
Fig. 7 Room temperature photoluminescence spectrum of ZnO nanorods in air (black line) and following 5min exposure in a 50 Torr ethanol atmosphere (red line). The excitation energy is 1 mJ/pulse.
The PL emission as a function of exposure time was tested for a number of ZnO overlaid LPGs corresponding to different growth times. A rather uniform behavior was revealed for LPGs bearing ZnO nanorods produced at medium to long growth times, namely 45 to 80 min. As observed in a 3-cycle experiment, immediately upon exposure of the sensor to ethanol the ZnO photoluminescence increases by as much as 30% over the corresponding value in air and remains so until the end of the 1st cycle when ethanol is flushed away (Fig. 8 ). Then a fast recovery is observed, however it appears, as judged by the PL intensity, that the material does not return to the initial, unperturbed state, signaling possibly slow desorption of ethanol molecules diffused well into the rods interior. Exposure of the sensor to ethanol in a 2nd and then a 3rd cycle leads to similarly sharp on/off transitions of the PL intensity corresponding to the beginning and end of each cycle.
Fig. 8 Response and recovery of the ZnO nanorods of photoluminescence signal changes ΔPL versus time for successive exposures to 50Torr ethanol atmosphere, for a sample fabricated employing 80min immersion in the chemical solution.
However the amplitude of the transitions gradually drops and this is an indication that the sensing operation becomes less efficient with repeated exposures, underlying the possibility of defect passivation. Passivation related changes for repetitive exposures have been also observed and interpreted in the optical measurements performed on the ZnO coated LPGs, but in a substantially lower extend.
4.3 Discussion on the underlying ethanol sensing mechanisms
The electronic mobility induced by ethanol into the ZnO nanorods result in conductivity changes which in turn are directly associated with optical absorption and refractive index changes. In previous studies, ZnO thin films have been used directly for the optical interrogation of ethanol vapors [16, 30] and butane [31] by exploiting guiding mode rejection waveguide schemes, detecting refractive index changes in the range of x10−5 and x10−3 respectively. Concerning ethanol sensing, studies have showed that ethanol adsorption on single crystal ZnO prisms [32] or on nanostructured thin films and nano-belts [12], resulted in a conductivity increase dependent upon the ethanol vapor concentration, and the operation temperature of the device. Moreover, Kuo, et al [33], found that ethanol adsorption on pristine and Fe-doped ZnO nanoparticles, resulted in reversible shortening of the Zn-O bond, which can be correlated with strain induced refractive index changes. Therefore, the adsorption of ethanol onto ZnO crystal structures induces both electronic and structural modifications, which in turn can affect other physical parameters of the material. We believe that both refractive index and optical absorption changes are involved and probed in the ethanol detection experiments reported herein. However, the effect each of those two physical quantities has individually on the spectral properties of the LPGs is of different magnitude and nature. While refractive index changes will affect both the strength and the spectral position of the LPG spectra, optical absorption changes will primarily affect the LPG strength [34]. This may constitute a basis for explaining the different trends of the LPG notch strength and wavelength position presented in Fig. 5(a), where greatest changes are observed in the LPG strength.
On the other hand changes observed in the PL intensity of ZnO nanorods in gas atmosphere, at high temperatures has been attributed to the chemisorption of oxygen on the nanorods surface and the subsequent oxidation reaction between the adsorbed oxygen and the detected gas [35]. In particular it is reported that when ZnO nanowires are exposed to air, an oxygen molecule (O2) adsorbs on the surface and forms an O- ion by capturing an electron from the conduction band, forming a depletion layer, which controls the density and the mobility of electrons in the nanorods [12, 35]. During the exposure to ethanol, which is a reducing gas, the captured electrons are released back to the conduction band resulting in an increase of charge carrier concentration, which can radiatively recombine [36]. As a consequence adsorption of ethanol on the ZnO surface induces an increase of the PL intensity as observed in the present PL studies (Fig. 7, 8). Upon exposure to air, the adsorbed ethanol molecules leave the surface of ZnO nanorods, allowing the de-trapping of electrons and the consequent drop of the PL signal intensity. The sensing behavior shown in Fig. 8, where the PL intensity is stable in ethanol atmosphere and decreases in air is well explained by the above mechanism, implying further that a constant number of molecules remains attached on the ZnO nanorods surface.
An important question arising is how to correlate the results obtained in the PL measurements with those obtained following optical interrogation of the ZnO overlaid LPGs. LPG optical measurements interrogate modifications primarily correlated with conductivity and structural changes (optical absorption and then refractive index changes) being localised primarily in the volume and less on the surface of the nanorods, but the photoluminescence integrates the behavior of all emissive defects and centres that can be excited by the ethanol vapors, especially those located at the ZnO surface. Therefore, these two measurement approaches do not lead to similar rise and decay trends upon exposure to ethanol, as this has been observed here but also by others [23]. Specifically, Comini, et al have performed NO2 and ethanol sensing experiments at room temperature, correlating conductivity changes with those of PL, wherein PL changes reached saturation at much faster rates [23]. Such interpretation can also describe the experimental results presented here, in which optical measurements probing optical loss and refractive index changes exhibit rise times of tens of minutes before reaching saturation, while PL measurements can substantially faster in response. PL measurements are likely to be correlated with ethanol activated centres located at the surface of the nanorods, thus exhibiting a fast time response. As aforementioned, LPG optical measurements refer to electronic and structural changes that take place into the volume of the nanorods. In this last scenario a diffusion law is more plausible, wherein longer penetration times are required for inducing significant refractive and optical density changes inside the ZnO nanorods [37], rendering the temporal response of the ethanol probe dependent upon film thickness and temperature.
5. Conclusions
We have presented a room temperature ethanol vapor sensing probe utilizing ZnO, in the form of nanorods, as a chemosorptive material overlaid onto long period fiber gratings. The manifold study performed herein illustrates several parameters determining the sensing capabilities of such a hybrid, optical fibre/nanostructured oxide sensing probe: the ZnO nanorod integration process onto the LPG and the real-time monitoring of the growth process, the optical sensing response of the developed probes versus fabrication conditions, as well as, the possible underlying physical mechanisms with respect to photoluminescence behavior, and their correlation with the optical fibre measurements. The above studies reveal that ZnO growth conditions, which define the opto-geometrical properties of the random nano-rods overlayer, dominate the spectral response of the long period grating device. This strong dependence was confirmed by investigating the ethanol response of ZnO overlaid LPGs under different growth conditions, but also by correlating their spectral behavior with existing simulation models. We are continuing our investigations in establishing a better and more consistent understanding of the ethanol interaction mechanism with ZnO nanorods, trying to quantify the optical changes induced. Preliminary results on the sensitivity of the ZnO overlaid LPG probe presented here, using gasoline vapors have been also obtained.
Acknowledgments
The authors are grateful to A, Manoussaki for carrying out SEM studies, and Dr. G. Kenanakis (FORTH-IESL) for fruitful discussions. This work was supported in part by the EC, Project SP4-Capacities “IASIS” No 232479.
References and links
1. L. Bansal and M. El-Sherif, “Intrinsic optical-fiber sensor for nerve agent sensing,” IEEE Sens. J. 5(4), 648–655 (2005). [CrossRef]
2. O. S. Wolfbeis, “Fiber-optic chemical sensors and biosensors,” Anal. Chem. 78(12), 3859–3874 (2006). [CrossRef] [PubMed]
3. S. Korposh, S. W. James, S.-W. Lee, S. Topliss, S. C. Cheung, W. J. Batty, and R. P. Tatam, “Fiber optic long period grating sensors with a nanoassembled mesoporous film of SiO2 nanoparticles,” Opt. Express 18(12), 13227–13238 (2010). [CrossRef] [PubMed]
4. J. Zhang, X. Tang, J. Dong, T. Wei, and H. Xiao, “Zeolite thin film-coated long period fiber grating sensor for measuring trace chemical,” Opt. Express 16(11), 8317–8323 (2008). [CrossRef] [PubMed]
5. A. Cusano, P. Pilla, L. Contessa, A. Iadicicco, S. Campopiano, A. Cutolo, M. Giordano, and G. Guerra, “High-sensitivity optical chemosensor based on coated long-period gratings for sub-ppm chemical detection in water,” Appl. Phys. Lett. 87(23), 234105 (2005). [CrossRef]
6. Z. Gu, Y. Xu, and K. Gao, “Optical fiber long-period grating with solgel coating for gas sensor,” Opt. Lett. 31(16), 2405–2407 (2006). [CrossRef] [PubMed]
7. M. Konstantaki, S. Pissadakis, S. Pispas, N. Madamopoulos, and N. A. Vainos, “Optical fiber long-period grating humidity sensor with poly(ethylene oxide)/cobalt chloride coating,” Appl. Opt. 45(19), 4567–4571 (2006). [CrossRef] [PubMed]
8. A. Kolmakov, Y. Zhang, G. Cheng, and M. Moskovits, “Detection of CO and O2 using Tin Oxide nanowire sensors,” Adv. Mater. (Deerfield Beach Fla.) 15(12), 997–1000 (2003). [CrossRef]
9. K. S. Yoo, S. H. Park, and J. H. Kang, “Nano-grained thin-film indium tin oxide gas sensors for H2 detection,” Sens. Actuators B Chem. 108(1-2), 159–164 (2005). [CrossRef]
10. G. Kenanakis, D. Vernardou, E. Koudoumas, G. Kiriakidis, and N. Katsarakis, “Ozone sensing properties of ZnO nanostructures grown by the aqueous chemical growth technique,” Sens. Actuators B Chem. 124(1), 187–191 (2007). [CrossRef]
11. W.-Y. Wu, J.-M. Ting, and P.-J. Huang, “Electrospun ZnO nanowires as gas sensors for ethanol detection,” Nanoscale Res. Lett. 4(6), 513–517 (2009). [CrossRef] [PubMed]
12. Q. Wan, Q. H. Li, Y. J. Chen, T. H. Wang, X. L. He, J. P. Li, and C. L. Lin, “Fabrication and ethanol sensing characteristics of ZnO nanowire gas sensors,” Appl. Phys. Lett. 84(18), 3654–3656 (2004). [CrossRef]
13. N. A. Yebo, P. Lommens, Z. Hens, and R. Baets, “An integrated optic ethanol vapor sensor based on a silicon-on-insulator microring resonator coated with a porous ZnO film,” Opt. Express 18(11), 11859–11866 (2010). [CrossRef] [PubMed]
14. D. Anglos, A. Stassinopoulos, R. N. Das, G. Zacharakis, M. Psyllaki, R. Jakubiak, R. A. Vaia, E. P. Giannelis, and S. H. Anastasiadis, “Random laser action in organic/inorganic nanocomposites,” J. Opt. Soc. Am. B 21(1), 208–213 (2004). [CrossRef]
15. N. N. Lathiotakis, A. N. Andriotis, and M. Menon, “Codoping: a possible pathway for inducing ferromagnetism in ZnO,” Phys. Rev. B 78(19), 193311 (2008). [CrossRef]
16. C. de Julián Fernández, M. G. Manera, G. Pellegrini, M. Bersani, G. Mattei, R. Rella, L. Vasanelli, and P. Mazzoldi, “Surface plasmon resonance optical gas sensing of nanostructured ZnO films,” Sens. Actuators B Chem. 130(1), 531–537 (2008). [CrossRef]
17. W. Zhang, “Automotive fuels from biomass via gasification,” Fuel Process. Technol. 91(8), 866–876 (2010). [CrossRef]
18. C. Liewhiran, A. Camenzind, A. Teleki, S. E. Pratsinis, and S. Phanichphant, “High performance ethanol sensor for control drunken driving based on fame-made ZnO nanoparticles,” in Nano/Micro Engineered and Molecular Systems,2007. NEMS '07. 2nd IEEE International Conference on, 2007, 672–677.
19. D. W. Lachenmeier, R. Godelmann, M. Steiner, B. Ansay, J. Weigel, and G. Krieg, “Rapid and mobile determination of alcoholic strength in wine, beer and spirits using a flow-through infrared sensor,” Chem. Cent. J. 4(1), 5 (2010). [CrossRef] [PubMed]
20. A. Klini, A. Mourka, V. Dinca, C. Fotakis, and F. Claeyssens, “ZnO nanorod micropatterning via laser-induced forward transfer,” Appl. Phys., A Mater. Sci. Process. 87(1), 17–22 (2007). [CrossRef]
21. L.-Y. Shao, J. P. Coyle, S. T. Barry, and J. Albert, “Anomalous permittivity and plasmon resonances of copper nanoparticle conformal coatings on optical fibers,” Opt. Mater. Express 1(2), 128–137 (2011). [CrossRef]
22. M. Konstantaki and S. Pissadakis, “Optically tunable long period fiber gratings utilizing a photochromic out-cladding overlayer,” Opt. Fiber Technol. 17(3), 168–170 (2011). [CrossRef]
23. E. Comini, C. Baratto, G. Faglia, M. Ferroni, and G. Sberveglieri, “Single crystal ZnO nanowires as optical and conductometric chemical sensor,” J. Phys. D Appl. Phys. 40(23), 7255–7259 (2007). [CrossRef]
24. C. Baratto, S. Todros, G. Faglia, E. Comini, G. Sberveglieri, S. Lettieri, L. Santamaria, and P. Maddalena, “Luminescence response of ZnO nanowires to gas adsorption,” Sens. Actuators B Chem. 140(2), 461–466 (2009). [CrossRef]
25. Y. Wang, Z. Zhou, Z. Yang, X. Chen, D. Xu, and Y. Zhang, “Gas sensors based on deposited single-walled carbon nanotube networks for DMMP detection,” Nanotechnology 20(34), 345502 (2009). [CrossRef] [PubMed]
26. A. Cusano, A. Iadicicco, P. Pilla, L. Contessa, S. Campopiano, A. Cutolo, and M. Giordano, “Mode transition in high refractive index coated long period gratings,” Opt. Express 14(1), 19–34 (2006). [CrossRef] [PubMed]
27. M. Nagao and T. Morimoto, “Adsorption of alcohols on zinc oxide surfaces,” J. Phys. Chem. 84(16), 2054–2058 (1980). [CrossRef]
28. S. W. James and R. P. Tatam, “Optical fibre long-period grating sensors: characteristics and application,” Meas. Sci. Technol. 14(5), R49–R61 (2003). [CrossRef]
29. A. B. Djurisić and Y. H. Leung, “Optical properties of ZnO nanostructures,” Small 2(8-9), 944–961 (2006). [CrossRef] [PubMed]
30. A. O. Dikovska, P. A. Atanasov, A. T. Andreev, B. S. Zafirova, E. I. Karakoleva, and T. R. Stoyanchov, “ZnO thin film on side polished optical fiber for gas sensing applications,” Appl. Surf. Sci. 254(4), 1087–1090 (2007). [CrossRef]
31. A. O. Dikovska, P. A. Atanasov, T. R. Stoyanchov, A. T. Andreev, E. I. Karakoleva, and B. S. Zafirova, “Pulsed laser deposited ZnO film on side-polished fiber as a gas sensing element,” Appl. Opt. 46(13), 2481–2485 (2007). [CrossRef] [PubMed]
32. W. Mokwa, D. Kohl, and G. Heiland, “Decomposition of ethanol and acetaldehyde on clean ZnO prism and oxygen faces,” Surf. Sci. 117(1-3), 659–667 (1982). [CrossRef]
33. G.-H. Kuo, H. P. Wang, H. H. Hsu, J. Wang, Y. M. Chiu, C.-J. G. Jou, T. F. Hsu, and F.-L. Chen, “Sensing of ethanol with nanosize Fe-ZnO thin films,” J. Nanomater. 2009, 316035 (2009). [CrossRef]
34. X. Daxhelet and M. Kulishov, “Theory and practice of long-period gratings: when a loss becomes a gain,” Opt. Lett. 28(9), 686–688 (2003). [CrossRef] [PubMed]
35. X. Chu, T. Chen, W. Zhang, B. Zheng, and H. Shui, “Investigation on formaldehyde gas sensor with ZnO thick film prepared through microwave heating method,” Sens. Actuators B Chem. 142(1), 49–54 (2009). [CrossRef]
36. D. Valerini, A. Cretì, A. P. Caricato, M. Lomascolo, R. Rella, and M. Martino, “Optical gas sensing through nanostructured ZnO films with different morphologies,” Sens. Actuators B Chem. 145(1), 167–173 (2010). [CrossRef]
37. G. Sakai, N. Matsunaga, K. Shimanoe, and N. Yamazoe, “Theory of gas-diffusion controlled sensitivity for thin film semiconductor gas sensor,” Sens. Actuators B Chem. 80(2), 125–131 (2001). [CrossRef]
