Silicon nanowire-based devices for gas-phase sensing

Abstract

Since their introduction in 2001, SiNW-based sensor devices have attracted considerable interest as a general platform for ultra-sensitive, electrical detection of biological and chemical species. Most studies focus on detecting, sensing and monitoring analytes in aqueous solution, but the number of studies on sensing gases and vapors using SiNW-based devices is increasing. This review gives an overview of selected research papers related to the application of electrical SiNW-based devices in the gas phase that have been reported over the past 10 years. Special attention is given to surface modification strategies and the sensing principles involved. In addition, future steps and technological challenges in this field are addressed.

Figure 1.

Simplified schematics of the SiNW-based (a) resistor and (b) SiNW-based FET to illustrate the differences in the electrical configuration and the way the nanowires are orientated with respect to the electrodes (E) and the source (S) and drain (D).

Figure 2.

Electrical characterization of nanowire thin-film transistors on plastic. (a) Schematic illustration of the active area of a transistor, with the electrodes and various layers labeled; (b) Scanning Electron Microscope image of the sensor platform. Each device (horizontal strip) is contacted by two Ti electrodes (oriented vertically) that extend to larger pads (top and bottom image edges). Inset: Digital photograph of the flexible sensor chip; (c) Electrical response of a nanowire-based sensor to 20 ppm (red curve), 2 ppm (blue curve), 200 ppb (green curve) and 20 ppb (black curve) NO₂ diluted in N₂. The gas is introduced to the sensing chamber after 1 min of flowing N₂. Inset: An extended response of the sensor to 20 ppb NO₂; the gas is introduced after 20 min of flowing N₂. This figure is composed of figures taken from [31], reprinted with permission from the Nature Publishing Group ©.

Figure 3.

(a) A schematic representation of the interdigitated structure used as a substrate to grow SiNWs. W = 200 μm, L = 3–21 μm, h =100 nm, SEM images of (b) the final chemoresistive sensor and (c) SiNW network detail onto the interdigitated structure; (c) The inset shows a SEM image of self-welding NW-NW junctions present in the SiNW network. This figure is composed of figures taken from [39], reprinted with permission from the American Institute of Physics ©.

Figure 4.

(a) Schematic of the SBNWFET, consisting of p-type doped nanowires; (b) Energy band diagram before H₂ is introduced, where E_C and E_V stand for the energy levels of the conduction and valence band, respectively. The holes do not experience a barrier and the contact is Ohmic. The native SiO₂ layer is omitted in this diagram because it is too thin to affect transport; (c) Energy band diagram after H₂ is introduced, showing the formation of a Schottky barrier; (d) Comparison of the responses of Ni and Pd contacted NWFETs to 1% H₂. This figure is composed of figures taken from [36], reprinted with permission from Elsevier B.V. ©.

Figure 5.

(a) Real-time electrical response curve of a device with vertical-standing SiNWs coated with a 7 nm thick Pd film to varying H₂ concentrations in air at room temperature. The inset shows a clear and reversible response behavior even at very low H₂ concentrations down to 5 ppm; (b) A plot of the sensitivity vs. H₂ concentrations revealing two regimes with different rates of sensitivity change; (c) A proposed model of the hydrogen-sensing mechanism. Panel (c1) represents the initial devices with Pd-coated SiNWs with two on-top electrodes as indicated by the yellow lanes. The inset of the left panel shows a magnified distribution of SiNWs inside a cluster, indicating that the wires do not touch each other. Panel (c2) illustrates nanowire contacts inside clusters with gaps between neighboring clusters. In this case, the current flows through slanted nanowires between clusters. The inset is a magnified picture of contacted distribution of SiNWs inside a cluster. Panel (c3) illustrates the formation of current paths between neighboring clusters by large volume expansion of the Pd film caused by absorption of high concentrations of H₂. This figure is composed of figures taken from reference [37], reprinted with permission from The Royal Society of Chemistry ©.

Figure 6.

(a) Schematic description of surface modification by self-heating of a nanowire: in method 1, nanoparticles are formed by a hydrothermal reaction via Joule heating of a SiNW. In method 2, a metal thin film is locally deposited on a SiNW after PMMA decomposition, metal evaporation and lift-off. SEM images of the nanoparticle-decorated SiNW via Joule heating in a liquid metal precursor environment; (b) Pd nanoparticles selectively coated on the heated SiNW; and (c) high magnification SEM image of Pd nanoparticles on the heated SiNW. This figure is composed of figures taken from [40], reprinted with permission from the Royal Society of Chemistry ©.

Figure 7.

(a) Schematic representation of the (left) APTES- and (right) peptide-modified SiNW surface. Note that two different peptides have been used, i.e., an ammonia recognition peptide and an acetic acid recognition peptide; (b) Electrical responses of the three different devices (APTES, NH₃ peptide and acetic acid peptide) to ammonia and acetic acid vapors (100 ppm in N₂) introduced to the sensing chamber after 5 min of flowing N₂; (c) Conductance responses of the peptide-nanowire hybrid sensors, averaged over a 5 min time window of target vapor exposure (starting 10 min after the analyte gas exposure), and normalized to the amine-terminated sensor. Figures b,c are taken from [44], reprinted with permission from American Chemical Society ©.

Figure 8.

(a) Schematic illustration of the periodically porous top electrode (PTE) nanowire array sensor concept; (b) Sensor response to various concentrations of NO₂ and NH₃ following 2 min of clean air: 1 ppm of NH₃ (red), 500 ppb of NH₃ (black), 1 ppm of NO₂ (green) and 500 ppb of NO₂ (blue) at ∼30% RH. This figure is composed of figures taken from [48], reprinted with permission from IOP Publishing Ltd ©.

Figure 9.

(a) The response of an APTES-functionalized SiNW device to (red) 5 μM solutions and (blue) 5 nM solutions. The inset shows the two different kind of interactions between TNT and the NH₂-terminated SiNWs; (b) Molecular structures of six different N-containing compounds: (1) TNT; (2) 2,6-dinitrotoluene; (3) 2,4-dinitrophenol; (4) aniline; (5) (1,3,5-trinitroperhydro-1,3,5-triazine; and (6) p-nitrophenol. Figure based on graphs taken from [55] and used with permission from Wiley-VCH©.

Figure 10.

(a) Molecular structure of DPCP and two structurally related phosphates; (b) Sensitive receptor towards DPCP. Compound 1 converts into compound 2 upon exposure nerve agent simulant DPCP. The red oval highlights an alkyne group, which can react with H-terminated Si to form a stable Si-C bond; (c) Change in drain current of a SiNW FET device that has been modified with compound 1 upon exposure to DPCP, which was introduced at t = 240 s. Figure c is taken from [54] with permission Wiley-VCH.

Figure 11.

Simplified scheme of the attachment of hexyltrichlorosilane (HTS) to the SiO₂ surface: (a) Preparation of surface hydroxyl (Si–OH) groups; (b) Exposure to trimethylamine (TMA) to form a hydrogen bond with the Si–OH group to make the oxygen atom more nucleophilic; (c) Exposure to 1.5 mM of HTS in chloroform. At this stage, the oxygen atom of the Si–OH group attacks the silicon atom of HTS to form a Si–O–Si bond as given in (d); (e) The presence of water residues assists the replacement of chlorine atoms by OH groups to form the final product given in (f); (g) Schematic illustration of molecular layer with different functional groups on the SiNW surface. Figure a–f and Figure g are taken from [63] and [64], respectively, reprinted with permission from American Chemical Society ©.