In a chemical sensor the active surface area is one of the important factors determining the detection limits or sensitivity. Due to their large surface-to-volume ratio, NWs are well suited for direct measurement of changes in their optical or electrical properties (e.g. PL, conductance/resistance, impedance) when exposed to various analytes. Furthermore, in the case of GaN, its chemical inertness and capability of operating in extreme environments (high temperatures, presence of radiation, extreme pH levels) is highly desirable for sensor design.
Chen et al. (2009) and Gonzalez-Posada et al. (2012b) have reported the dependence of the conductivity, photocurrent, and photocurrent decay time of single GaN NWs on the ambient conditions, notably on the presence of oxygen (Chen et al. 2009; Gonzalez-Posada et al. 2012b) and hydrogen (Chen et al. 2009). The results are comparable even if the works describe the performance of m-oriented CVD-deposited NWs (Chen et al. 2009) and (000-1)-oriented PAMBE-grown NWs (Gonzalez-Posada et al. 2012b). An enhancement of the chemical sensitivity of GaN NWs by insertion of an AlN barrier has been proposed by den Hertog et al. (2012). The photocurrent response of single GaN NWs to UV illumination in vacuum, air, and pure oxygen at room temperature are illustrated in Fig. 9.16. As a function of the measuring environment, two major differences are identified: (i) the average steady-state photocurrent measured in vacuum is higher than the one in air/oxygen; (ii) the transient photocurrent shows that the photoresponse in air or oxygen is faster than in vacuum. These observations are assigned to adsorbate-induced variations of the surface band bending and carrier lifetime.
Although these effects can find applications in the domain of chemical sensors, the specificity of the surface/adsorbate interactions is limited. The idea of functionalizing or decorating the NW surface with metal or metal-oxide nano-particles or nano-clusters aims at solving this deficiency.
Teubert et al. (2011) and Wright et al. (2009) have studied the Pt- functionalization of GaN NWs for detection of H2, measuring chemically induced variations of the NW PL or the NW resistance, respectively. With this purpose, NW samples were coated with a Pt layer of a nominal thickness of 5-7 nm, which spontaneously arranges forming Pt islands. Figure 9.17 plots the PL response of GaN NWs with embedded GaN/AlGaN nanodisks to H2 and O2, comparing bare surfaces and NWs functionalized by deposition of Pt. Hydrogen-induced dipole fields can suppress the surface recombination in Pt-coated NWs leading
Fig. 9.16. (a) Different photocurrent responses of GaN NWs to 325-nm UV excitation measured in vacuum, air, and pure oxygen. The shadow shows the duration with photoexcitation. (b) Normalized photocurrent rise curves in vacuum, air, and pure oxygen. Inset: A schematic of the electron-hole spatial separation mechanism induced by surface band bending in a GaN NW. SDR denotes surface depletion region and NR denotes neutral region. (c) Normalized photocurrent decay curves in vacuum, air, and pure oxygen. (Reprinted, with permission from Chen et al. (2009), © 2009 American Institute of Physics.)
Fig. 9.17. Transient response of the relative PL intensity from AlGaN nanodisks inserted in GaN NWs to variations of the gaseous environment (T = 150°C). In the presence of hydrogen (H2 concentration of 0.1% in synthetic air) an increase of the PL intensity for Pt-coated NWs is observed. In contrast, a decrease of the PL signal is detected in the presence of oxygen (O2 concentration of 0.1% in nitrogen) which can be suppressed by Pt coverage. (Reprinted, with permission, from Teubert et al. (2011), © 2011 IOP Publishing Ltd.)
to an enhanced PL intensity in hydrogen-containing atmosphere. For uncoated NWs, the oxygen-induced enhancement of the surface recombination induces a reduction in PL intensity.
Regarding the effect on the NW resistance, non-linear relative variations of «1.7% and «1.9% are obtained when exposing the Pt-coated NWs to 200 ppm and 2000 ppm of H2 in N2, respectively. Better results are obtained by replacing Pt with Pd, which results in relative responses of «7.4% at 200 ppm and «9.1% at 1500 ppm of H2 in N2 (Lim et al. 2008). More recently, relative responses of 34.1% at 100 ppm H2 were achieved by anealing the Pd-coated GaN to form Ga2Pd5 nanodots (Kim et al. 2011).
Aluri et al. (2011) demonstrated highly selective and sensitive sensors using GaN NWs decorated with TiO2 nanoclusters. Hybrid sensor devices were developed by fabricating two-terminal devices using single GaN NWs followed by the deposition of TiO2 nanoclusters using RF magnetron and annealing at 700° C. The catalytic properties of TiO2 enables selective sensing of aromatic compounds with additional selectivity for methyl group substitution, i.e. these sensors can distinguish toluene from other aromatic compounds. In the proposed device the reaction of the analyte is measured as a variation of the photocurrent under UV excitation. Thus, a change of photocurrent was observed when the sensors are exposed to the vapors of aromatic compounds (toluene, benzene, ethylbenzene, and xylene mixed with air), while they had no response to non-aromatic organic compounds such as methanol, ethanol, isopropanol, chloroform, acetone, and 1,3-hexadiene. The sensitivity range for the noted aromatic compounds was from 1% down to 50 ppb at room temperature. The authors speculate that photoinduced oxygen desorption and subsequent capture of holes by organic adsorbate molecules on the surface of TiO2 clusters produces a local gating effect, which is responsible for the sensing action.
By addition of sputtered Pt, single GaN/(TiO2-Pt) NWs become only sensitive to methanol, ethanol, and hydrogen, whereas higher carbon-containing alcohols (n-propanol, iso-propanol, n-butanol) did not produce any sensor response (Aluri et al. 2012). Reference single GaN/Pt NWs were only sensitive to hydrogen, and not to methanol or ethanol. Figure 9.18) shows the comparative summary of the sensing behavior of GaN/TiO2 NWs, GaN/(TiO2-Pt) NWs and GaN/Pt NWs to 1000 ppm of various analytes in air. The sensitivity is defined as the relative variation of the single NW resistance when exposed to the analyte,
i.e. Rgas-Rair/Rair, where Rgas and Rair are the resistances of the sensor in the presence of the analyte-air mixture and in the presence of air only, respectively (Rair is replaced by Rnitrogen for H2 sensing experiments).
On the other hand, semiconductor NWs, having a diameter comparable to biologically-relevant molecules, are interesting nanostructures for biosensing
Fig. 9.18. Comparative sensing behavior of functionalized GaN NWs to 1000 ppm of various analytes in air: light grey bar graphs represent single GaN/TiO2 NWs, light and dark grey bars represent single GaN/(TiO2-Pt) NWs, and the dark grey plot represents single GaN/Pt NWs. Other chemicals which did not produce any response in any one of the sensors are not included in the plot. The zero line is the baseline response to 20 sccm of air and N2. The error bars represent the standard deviation of the mean sensitivity values for every chemical computed for five different devices with diameters in the range of 200-300 nm. (Reprinted with permission from Aluri et al. (2012), © 2012 IOP Publishing Ltd.) devices (Cui et al. 2001). Recent studies have demonstrated successful application of GaN NWs as transducers in highly sensitive label-free DNA-sensing, using cyclic voltammetry, electrochemical impedance spectroscopy, and PL techniques (Chen et al. 2009b; Ganguly et al. 2009). The DNA-immobilized GaN NWs are found to possess distinct Faradaic characteristics compared to the unmodified NWs. Based on these results, Chen et al. (2011b) have demonstrated label-free identification of specific DNA sequences by connecting a sample of GaN NWs synthesized on Si by Au-catalyzed CVD to the gate of a commercial ra-MOSFET, reaching a detection limit in the level of 10~18 molar, about six orders of magnitude lower than that of GaN 2D layers in the same configuration.