Drifts, chemical sensors

Higher-order chemical sensing can alleviate some inherent problems of chemical sensors. For example, a sensor array can mathematically correct for systematic drift. It also provides cross selectivity for elimination of interference. 

Sensor drift is a first serious impairment of chemical sensors. They alter over time and so have poor repeatability since they produce different responses for the same odour. That is particularly troublesome for electronic noses (Remain et al. 2002). The sensor signals can drift during the learning phase (Holmberg et al. 1997). To try to compensate the sensor drift, three types of solutions were tested for our applications. 

Figure 3.25 Autocorrelation function for a time series with drift, as found for measurements with a chemical sensor.

Significant efforts have been made to develop and test new metal oxides. However, the application of chemical sensors still faces problems such as selectivity and long-term drift due to stoichiometry changes and coalescence of crystallites. The notion of preparing multipurpose devices has been replaced by the development of sensors tailored for specific and focused applications. 

Since the dawn of chemical sensors, there were high hopes of rapid and widespread utilization of these devices. This predicted success, however, has not yet been achieved, due to several causes (bio)chemical fouling of the sensor, instability in the sensor signal resulting in drift, and lack of available selector materials, specific for all species to be detected. 

The presence of flexible side-chains make them soluble in common organic solvents [98,204-7]. These polymers have been proved by solid-state NMR and small-angle X-ray diffraction measurements to have hexagonal columnar structures in which the phthalocyanine cores are horizontally stacked with respect to the columnar axis, even at ambient temperatures [107,139,207]. The rigid rod nature of these polymers have been proved experimentally [207,208]. Their completely hydrophobic polymer allows for the controlled and highly reproducible deposition of ultra-thin films. LB films are used as branes in field effect transistors (PET S) and chemical sensors [209]. Devices based on the LB films show stable electroactivity and the electrodes show an almost Nemstian response. The change of pH in the electrolyte does not show any long time drifts. 

Fouling of the pH sensor may occur in solutions containing surface-active constituents that coat the electrode surface and may result in sluggish response and drift of the pH reading. Prolonged measurements in blood, sludges, and various industrial process materials and wastes can cause such drift. Therefore, it is necessary to clean the membrane mechanically or chemically at intervals that are consistent with the magnitude of the effect and the precision of the results requited. 

Zuppa et al.60 have used SOMs in the assessment of data from an electronic nose. Six chemicals—water, propanol, acetone, acetonitrile, butanol, and methanol—were presented at varying concentrations to a 32-element conducting polymer gas sensor array. The output was used to train a group of SOMs, rather than a single SOM, to avoid the problems of parameter drift. One SOM was associated with each vapor, and with suitable use of smoothing filters, the SOM array was found to perform effectively. 

Once the device has been fabricated, the chemically sensitive coating applied, and the supporting RF electronics turned on, acoustic-wave sensors sometimes do not perform as expected or desired. Drift may be excessive sensitivity or selectivity may be inadequate to solve the immediate problem. Section 6.5 outlines several popular strategies for enhancing AW sensor performance through careful system design. 

This section will broadly describe some of the practical system design strategies that are being used to minimize problems that sometimes plague all types of chemical vapor sensors such as baseline drift, inadequate sensitivity, and inadequate selectivity. While special emphasis will be given to acoustic sensors, the approaches described here are generally applicable to any vapor sensing device. 

Overall, the addition of a vapor-concentrator device to a chemical vapor sensor can produce dramatic enhancements in performance. Significantly lower vapor concentrations can be reliably detected from the combined effects of source concentration enrichment, which increases the apparent sensor signal, and baseline drift compensation, which reduces the apparent noise produced by the sensor. Unfortunately, these performance enhancements come with a fairly heavy price in the form of additional pumps, valves, traps, and increased energy consumption requirements. 

Optical interferometer techniques can achieve resolution far beyond 1 nm. Because of this high sensitivity of the interferometer sensor, the direct detection of small molectdes at low concentrations should be possible [15]. Detection is generally limited by electronic and mechanical noise, thermal drift, light source instabilities, and chemiced noise. But interferometric devices have an intrinsic reference channel which ofiers the possibility of... 

Several challenges remain for the ultimate practical use of these sensors. The response time of the solid state sensors are short (seconds) for initial sensing, but recovery times range from minutes to hours at room temperature. The stability of the sensor to drift associated with accumulation of fixed charge at interfaces, as well as the high sensitivity to ubiquitous urban pollutants ozone and N02 are problematic. All MPc OTFTs show some response to moisture, and conductivity is also temperature sensitive so that humidity and temperature compensation are essential. On a basic research level, the detailed characterization of charge trapping states, electronic structure, and the interactions with analytes is not yet fully understood on a quantitative theoretical basis. The time response of sensor initiation and recovery is also not understood in a detailed manner. In spite of these limitations, the intrinsic chemical stability of MPc compounds and their compatibility with microsensor array fabrication make these candidate OTFTs for further research and development. 

---