We explore the use of evanescent wave cavity ring-down spectroscopy (BW-CRDS) for water detection through a signal-to-noise ratio analysis. Cavity ring-down spectroscopy (CRDS) is an emerging optical absorption technique that employs the mean photon decay time ofa high-finesse optical cavity as the absorption-sensitive observable. EW-CRDS is a novel implementation of CRDS that extends the technique to surfaces, films, and liquids by employing optical cavities which incorporate at least one total-internal-reflection (TIR) mirror. The concomitant evanescent wave is then used to probe the absorption ofan ambient medium at the TIR surface also through a change in the photon decay time. By employing miniature monolithic cavities with ultra-smooth surfaces that are fabricated from ultra-high transmission materials, extreme sub-monolayer detection sensitivity is readily achieved. The detection of water by EW-CRDS with a fused-silica resonator provides an interesting and important application, since the nascent hydroxylated Si02 surface is expected to show a high natural affinity for adsorption ofwater through hydrogen-bonding interactions. Furthermore, in the 13 80 nm spectral region where water absorbs strongly, low-OH-content fused silica has extremely high bulk transmission. These factors potentially provide the basis for a novel water sensor.

Photothermal interferometry has been demonstrated as a technique that can detect vapors with extremely high sensitivity (parts-per-trillion levels). Our present research uses a photothermal detection scheme that incorporates tunable sources and a modified Jamin interferometric design to provide high selectivity and sensitivity for organo-phosphate vapor detection. Two possible tunable excitation sources are being studied for this sensor technology, a tunable CO2 laser and difference frequency mixing of a tunable NIR laser with a fixed wavelength NIR laser in a nonlinear crystal. The modified Jamin design imparts superior vibrational immunity by ensuring both interferometer beams encounter common optical elements. Examining the two complementary optical outputs of the interferometer, phase shifts on microradian levels have been detected. Trace chemical vapor detection is accomplished by introducing the tunable excitation laser source across the path of one interferometer beam providing a phase shift due to absorptive heating. Preliminary results indicated parts-per-billion level detection of both DMMP and DIMP using ~ 400mW of CO2 laser power at appropriate wavelengths.

Photoacoustic spectroscopy of gases is one of the most sensitive spectroscopic techniques available, often achieving part per million (ppm) or part per billion (ppb) sensitivity. This technique is usually performed by containing the gas sample within a cell coupled to an acoustic transducer. Imaging of photoacoustic response over an area requires scanning of the excitation beam. At the INEEL, research is underway to extend photorefractive dynamic holography to full field spectroscopic imaging without scanning. The photorefractive effect in Bismuth Silicon Oxide is exploited to demodulate the optical phase shift of a signal beam traversing the test gas and coincident with a tunable chopped excitation beam. Molecular absorption at the excitation wavelength produces heat that causes local expansion and subsequent acoustic wave radiation. A model of the photoacoustic absorption and optical phase detection process has been developed. Measurement and modeling results are presented that illustrate the ability of the method to detect water vapor and Hydrogen Fluoride concentrations in nitrogen atmosphere backgrounds near 800 nm, currently producing sensitivities in the 20-1000 ppm. Limitations of the technique and methods for extending to ppb sensitivities with infrared excitation are discussed.

High-resolution, miniature integrated spectrometers have been constructed for the NIR and MIR spectral ranges, based on MPBT's proprietary IOSPEC technology. Advanced slab-waveguide integrated optics have been employed to extend the performance of the miniature IR spectrometers to rival that of much larger FT-JR spectrometers. Monolithic integration of the miniature spectrometer, input optics and detector array provides a very compact and robust package that is suitable for industrial and field environments. Despite the compact size of the spectrometers, resolutions of 4 to 8 cm-1 are achievable over dedicated spectral ranges (2000 to 4000 nm, respectively). These spectrometers are coupled to 256channel linear detector arrays controlled by software based on Visual C++ to provide rapid spectral acquisition and analysis. This technology facilitates on-line infrared spectral analysis of an industrial or biochemical process at scan rates exceeding 200 spectra/sec. Since the spectral data is measured directly, significantly less data processing is required than for FT-JR techniques, allowing more CPU time for spectral identification and analysis. Multi-channel, time-resolved spectral measurements permit the study of the intermediate steps in a process or reaction. This paper discusses recent advances in the performance of the miniature integrated spectrometers. New detector geometries and data processing techniques have facilitated a substantial improvement in the overall system SNR over that feasible with typical sequentially-scanned detector arrays. Preliminary experimental transmittance spectra of optical filters and plastics are presented.

In the past decade interest has focused on the remote analysis of emissions from motor vehicles using spectrometric techniques, driven by recognition of the fact that a very large proportion of the environmental damage done in this way originates with a small percentage of vehicles. Several instrumentation manufacturers now market such devices, and others are developing new technologies that will offer opportunities for enhanced performance and lower cost. In this paper we review the evolution of technologies and methodologies applied to motor vehicle exhaust emissions, ranging from simple broadband sources and band-pass filters to tunable diode laser absorption spectroscopy. In so doing we examine the compromises and sources of error inherent in each which have earned such devices a very variable reputation, at least in the early years of development. We also look at techniques that may have the potential to solve these problem, and critically examine the reasons why these have not (yet) been applied. In conclusion, we will present initial findings and results from a European consortium studying the problems of costeffective emissions monitoring, and validation of emissions inventory data using complementary numerical modelling techniques.

The Modulation Transfer Function(MTF) is a very important performance parameter to an optical imaging system which is used as a remote sensor. Over a long period of time we applied ourselves to study the method to detect the MTF of an imaging system with single detector and continuous output. For an imaging spectrograph adopting linear array detectors with reading out circuit and with “discrete” output, it is more difficult to analyze and detect the MTF of such imaging system than that of the former. After deducing from the linear system theorem in this paper, an imaging spectrograph with linear array detectors and with multi-line scanning can be regarded as a low-passed filer plus a sampling process in its scanning direction. By adopting “over-sampling” method, using periodical spatial square wave function with the same frequency as the system’s Nyquist frequency as input, the MTFs of its different detectors to different wavebands were obtained. From the square wave response function of this system, we can easily find that there are the same peak values corresponding to the periodical square wave peak values, the only one difference among them is that they had different vale depths, that is to say their MTFs were different, so we can make a conclusion that the method to measure the MTF is feasible.

Water - general-purpose solvent and a important component of uncountable set of objects of the living and lifeless nature. The development of molecular and structural - chemical representations has allowed to give depleting explanation of a special talent of molecules of water to derivate connection with fragments almost of any matters. Began to be elucidated also role of "bonding" water in originating major physical characteristics of hydrated matters -clay, gypsum, of a cement rock some types of ferroelectric.At last, after discovery of L. Poling in 1961 of intercoupling between a phenomenon of a narcosis and crystallization of hydrates of drug matters became apparent, that the water, bound with biological molecules, somehow participates in control ofbiochemical processes. The main role in formation of the unified approach to immense variety of problems of "bonding" water belongs to to molecular aspect. At a level of molecules all world of connections of water - hydrates both solutions and their transmutations is operated by unified force concluded in combination of hydrogen bonds and ion - dipole interplays of molecules of water with fragments of other matters. The indicated rather gentle closely coupled interfaces are responsible for unique capacity of water to derivate connections with huge set of matters, stipulate a development of major properties and features of behavior of hydrates. To their number concern a crystallization, of solid phase transformations, including ferroelectric, partial or full dehydration. In aqueous solution the modification of the short-range order in arrangement of molecules can result in to phase change in a liquid -so-called stratifjing. The combination of stratifying and reallocating between nonniixing with kinetic effects conditioned by deboosting of Brownian motion and diffusive mass transfer near to a stagnation point, gives padding capabilities for usage in problems of a transmission of information with the help of chemical signals and allow with exclusive accuracy to regulate a carrier and high scale ofspecificity ofconcentrating in one offluid phases. In activity through of isotope WD exchange the movability of 'bonding' water in solutions of a D-glucose saccharose and tretbutanol is investigated. The isotopic exchange was conducted from a vapour phase at 1 00 aliquot excess of molecules D20. It is established, that in a system a D-glucose - tretbutanol - water, saccharose - tretbutanol - water the considerable proportion of molecules H20 is unapproachable for isotope H/D of exchange, that it is possible to explain by deboosting diffusive ofmass transfer near to a stagnation point. The deboosting of diffusion near to a stagnation point of a gravitational segregation, apparently, should result in to deboosting ofchemical reactions, and, therefore, is that in physics call non-linear, or control, member.

Molecular imprinting is a novel way to create sensitive layers for chemical sensors. Polymers can be moulded with analyte molecules, the template, which generate cavities in the polymer matrix that are capable of selectively incorporating the analyte. Polyurethanes were synthesized from aromatic monomer components in order to create sensitive layers for the detection of polycyclic aromatic hydrocarbons (PAHs). The large number of hydrophilic groups in these layers guarantees sufficient wetting of the coating. The cavities are often of slightly greater size than the imprinting molecule, resulting in higher sensitivity for analytes somewhat larger than the template. At elevated temperatures the higher reactivity of the monomers leads to a tighter fit of the polymer matrix around the template, thus increasing selectivity and sensitivity. In addition to polymerization conditions the innovative method of double imprinting, i.e. using two different templates, allows the variation ofthe nature and the ratio of the templates, which leads to better sensor effects. Combining these layers with selective detection methods such as fluorescence spectroscopy improves the selectivity ofthe sensor system even more. Even complex mixtures such as coffee can be characterized. Xanthine derivatives can be differentiated with mass-sensitive measurements using divinylbenzene-acrylic acid copolymers.

Commercial polymer microfiltration membranes were surface-modified with a graft copolymer of a functional monomer and a crosslinker in the presence of a template (triazine-herbicide). As result, membranes covered with a thin layer of imprinted polymer (MIP) selective to the template were obtained. The influence of the polymerization conditions on membrane recognition properties was studied by membranes’ capability to adsorb the herbicide from aqueous solution. The MIP synthesis is possible in both organic solvents and water. The low thickness of the MIP-layer and the porous membrane structure enable the highly specific sorption of the template at very high filtration rates. The possibility to introduce specific binding sites into porous membranes opens a general way to design new high-performance affinity membranes with application such as solid phase extraction.

Cyclohexapeptides were bound covalently to interference and gold layer and tested for recognition of amino acids in liquid phase. The peptides consist of three L -amino acids as variable building block. Attachment to the surface took place by reaction of the amino group in the side chain of L-cysteine. Reflectometric Interferance Spectroscopy (RIfS) was used for the determination of the interaction of aminoacids. Surface Plasmon Resonance (SPR) was used for validation and additional measurement with small peptides .Cyclopeptides with different side chains demonstrated distinct recognition pattern for different amino acids. The interaction was fast and reversible.

The intention for using microporous polymer is the enhancement of the selectivity and the sensitivity. Microporous polymers show the possibility through the control for analyte´s mobility into the matrix to extend the sensitivity spectrum for chemo sensors. A sensitive layer with molecular sieving properties is introduced in a sensor array, enhancing the discrimination capabilities of the sensing method tremendously. The volume and the size of the analytes were correlated with the sorption of the analytes in the glassy polymer UE2010. It was found that the size selective sensitive layers of the glassy polymer can discriminate effectively analytes, which are bigger or smaller than the mean pore size. The kinetic diameter ?kin of an adsorbate molecule is the molecular scale to correlate the sorption behaviour of a molecule with the aperture dimension of molecular sieves as zeolites or e.g. microporous polymers. ?kin is widely employed to describe kinetics, permeability and the permselectivity of adsorbate molecules.

Several types of selective materials are frequently used in chemical sensors such as natural antibodies, synthetic host substances (calixarenes, cyclodextrines, etc.) molecularly imprinted materials, or conventional polymers. For a systematic development of those materials, their sorption behavior for interesting analyte substances and potentially interfering compounds has to be thoroughly characterized, which can be a time-consuming and error-prone task. Moreover, using the respective sensor principle itself for this characterization an exact relation between the sensor signal and the underlying partition coeÆcient or sorption isotherm can often not be obtained. In this paper, we present an automated method for the direct determination of polymer/water partition coeÆcients of volatile organic compounds that consists of an automated uid handling system, a dedicated partitioning cell and a purge-and-trap gas chromatography (PT-GC) unit. The main application of this novel system is the characterization of layer materials for infrared evanescent wave spectroscopic (IR-EWS) sensors, however an extension to other problems is conceivable. The whole experimental procedure comprising calibration of the GC system, preparation of test solutions, analyte partitioning, sample analysis, as well as the necessary cleaning steps is performed automatically under computer control. Hence, this system can be operated unattendedly, yielding a reasonable throughput with comparatively low expenditure of human labor.

We have studied catalytic thin film resistors made from a Pd and Ni alloy, and propose a method for dramatically reducing the drift of the measured resistance. The resistances of Pd films increase monotonically when exposed to hydrogen, however a stable baseline is difficult to achieve and alpha to beta phase transitions result in hysteresis. It is known that at high hydrogen concentrations, the Pd film cracks and delaminates, however long-term exposures to low concentrations of hydrogen can also result in delaminations. Studies using Pd/Ni alloys show that the phase transition can be suppressed. High temperature anneals in 2 % hydrogen, and the addition of a Ti adhesion layer is shown to reduce drift. Usually long term studies on films are conducted in an ordinary air (oxidizing) atmosphere; however, we report here on studies carried out in a reducing atmosphere of 0.1% hydrogen in nitrogen for 6 months on two sensor structures, field effect transistors (FETs) and resistors. The Sandia Robust Hydrogen Sensor platform containing integrated heaters, temperature sensors, and hydrogen sensitive resistors and FETs was compared to a Sandia Wide Range Sensor containing a 10 atomic percent Ni/Pd (1000Å) alloy resistor with a (100Å) Ti adhesion layer. After six months the two hydrogen sensing resistors on the Robust platform, without an adhesion layer, read a hydrogen concentration of 61% and 2.3%, while the Wide Range Sensor read a hydrogen concentration of 0.102%, which is a dramatic improvement in limiting baseline drift.

In many chemical measurements, a sample must be transported from an inlet, to a measurement "cell" to a detector regardless of size of the measurement system. For micro-instruments that may be used in environmental monitoring applications, micro-fluidics provides a irreplaceable way of transporting gaseous or liquid micro-samples from inlet to outlet. Micro-channels and micro-cells can be fabricated using anisotropic wet chemical etching and crystalline Si (c-Si). Despite wide applicability ofthe approach, the mechanism of anisotropic wet chemical etching of c-Si is not well understood. In this paper, a proposed reaction mechanism for such etching is discussed in detail.

Control of the polymer surface chemistry is a crucial aspect in the development of plastic microfluidic devices. When commercially available plastic substrates are used to fabricate microchannels, differences in the electroosmotic flow (EOF) from plastic to plastic can be very high. Therefore, we have used polyelectrolyte multilayers (PEMs) to alter the surface of microchannels fabricated in plastics. The PEMs are easily fabricated and provide a means for controlling the flow direction and the electroosmotic mobility in the channels. Optimal modification of the microchannel surfaces was obtained by coating the channels with alternating layers of poly(allylamine hydrochloride) and poly(styrene sulfonate). The efficacy of the surface modification has been evaluated by measuring the electroosmotic flow mobility. When microchannels prepared in different polymer substrates were modified with PEMs, they demonstrated very similar electroosmotic mobilities. The PEMs have also been used to immobilize chemically selective molecules in the microchannels. In addition, relatively complex flow patterns, with simple arrangements of applied voltages, have been realized by derivatization of different arms of a single device with oppositely charged polyelectrolytes. Flow in opposite directions in the same channel is also possible; a positively derivatized plastic substrate with a negatively charged lid was used to achieve topbottom opposite flows.

Portable instruments offer several advantages in environmental mercury analysis including reduced sample handling and the possibility of obtaining results on-site in near real-time. This paper describes steps taken in our laboratory toward the goal of a portable shoebox-size spectrometer for measuring mercury in the environment.

Microfabricated devices are becoming increasingly important in biotechnology, especially in the areas of drug delivery, biological fluid analysis, and biological sensors. Interactions between fluid constituents and device surfaces become important when device dimensions reach the micrometer scale. We are reporting the synthesis and application of an octyl trichlorosilane ether of a short chain mono methylated poly(ethylene glycol) (PEG8,7) for control of protein adsorption on silicon surfaces. Self-assembled monolayers (SAMs) of the PEG8,7, a commercially available long chain poly(ethylene glycol) silane (PEG 5000), and 1H,1H,2H,2H,- perfluorooctyltrichlorosilane (13F) were prepared. These three SAM surfaces and uncoated silicon oxide were compared for control of protein adsorption of bovine serum albumin (BSA) under nonflow conditions. PEG8,7 was shown to be the best surface for inhibiting protein adsorption. The surfaces were analyzed using X-ray photoelectron spectroscopy (XPS) to quantify protein adsorption.

The air quality in the domestic area, workshops or industrial work places is of great interest, especially in respect to humidity and vapors. The simplest way to determine humidity is by resistive measurements using interdigital transducers (IDT). Combining these devices with polyelectrolytes, changes in conductivity due to variations of humidity can be monitored. A centralized surveillance of different control points can be accomplished by wirelessly, interrogative SAWs. A problem in measuring humidity by IDTs arises since cross sensitivities to solvent vapors are observed due to ion solvation effects. By varying the counterions from Li+ to Cs+ organic vapors are no longer able to compete with the hydration and selective moisture detection is achieved. Water insoluble sensor coatings which ensure long-term stability and resistance against flooding were synthesized by on-chip cross-linking on the surface of the applied microelectronic IDTs. Thus, a cross-linked Cs+ polyelectrolyte was designed as a selective sensor coating for humidity monitoring. One port wirelessly, interrogative SAW-devices have one transducer connected to an antenna and reflectors for identification (ID-tag). The interrogator transmits a burst signal and the sensor responds with a burst of signals depending on the number of reflectors. The ID-tags used have two reflectors which are approximately 4 and 1 5mm away from the transducer which respond in bursts at 3 and 9 ?s. On coating the area between the two reflectors with more hydrophilic or hydrophobic materials humidity or solvent vapors can be determined via the reflected amplitudes.

Although acoustic wave sensors that use piezoelectricity have many advantages, all of them need electrical connections to excite the piezoelectric crystals with an alternating voltage. This paper presents a new type of continuously operating, in-situ, and remotely monitored sensor that doesn't require electrical connections. The new sensor is comprised of a magnetoelastic metallic glass ribbon. Its sensing principle is similar to acoustic wave sensors. An externally applied alternating current (ac) magnetic field is used to excite magnetoelastic waves inside the magnetoelastic thin ribbon. Frequency responses are monitored with a pickup coil located outside the test area and the resonant frequencies are measured. The sensor responds to mass loading as a microbalance by decreasing its resonant frequency. When immersed in liquid, its resonant frequency is correlated with the square root of the product of liquid viscosity and density. We studied the relationship between the frequency shift and the square root of the product of viscosity and density of a starch solution. We found that the frequency shift was linearly proportional to the starch concentration. After bonding a poly-hydroxyethyl acrylate (poly-HEA) membrane to the magnetoelastic ribbon, the sensitivity of the sensor to water loading was greatly increased. The new sensor has also been used to monitor polymer curing. After bonding the ribbon with a pH sensing membrane, it was used to monitor pH. Because the sensor does not require electrical connections, it can remotely monitor concentrations in situ in a sealed container.

Applications like the warning about agricultural pest infestations, the detection of spoilt food during storage and transport as well as the monitoring of smoldering fires require highly selective odor sensing techniques. Insect antennae that have been optimized by evolution over million of years are most suitable for such a sensitive and selective detection of certain organic substances in air. The utilization of this highly specialized sense of smell from insects needs in terms of analytical tools, however, an adaptation of the antenna to the microelectronic technique. Therefore, a beetle/FET (field-effect transistor) interface as an innovative biosensor has been developed. This BioFET (biologically sensitive FET) is based on the direct combination of the intact chemoreceptor of an insect with the gate of a FET by means of an electrolyte solution. Depending on the experimental set-up, two different biosensor configurations, namely a whole-beetle BioFET and an isolated-antenna BioFET have been designed. In both configurations, the organic compound that is detected by the beetle initiates a recognition process at its nerve cell membranes, which results in a net potential over the whole insect antenna. Then, this potential drop modified the gate conductivity and consequently, the drain current of the FET. By applying various kinds of insect antennae (e.g. ofthe Colorado potato beetle and the steelblue jewel beetle) different odor concentrations, such as cis- 3-hexen-1-ol, guaiacol and 1-octen can be detected down to the low ppb range.

A novel system incorporating optical fiber long-period grating (LPG)-based sensors for rapid detection of biological targets is presented to address the current need for highly responsive, inexpensive, instrumentation for in-situ subsurface bioremediation technologies. With the appropriate configuration, the LPG sensor is able to measure key environmental parameters. The sensor allows for highly sensitive, real-time, refractive index measurements and by applying affinity coatings to the fiber surface, specific binding of molecules can be accomplished using swellable polymers or ligand-based affinity coatings. Advantages of the sensors have are that they are highly responsive, low profile, and can be serially multiplexed within a single-ended probe-like arrangement. This arrangement can be utilized either locally for site characterization or as a distributed sensor to map contaminant levels at multiple depths over a large area. The performance advantages make optical fiber sensors ideal for detection of environmental targets in drinking water, groundwater, soil, and other complex samples. This paper presents recent long-period grating-based sensor results that demonstrate the potential for bioremediation as well as a variety of other chemical and biological sensing applications.

A new class ofsensors has been designed and prepared based on replacing the original cladding material on a small section of an optical fiber with a conducting polymer or other environmentally sensitive material. Vapor induced chemical interactions with the polymer result in refractive index and optical absorption changes in the polymer cladding. These changes lead to an optical intensity modulation induced within the multi-mode optical fiber. Polyaniline and polypyrrole were used as the modified cladding material on the fiber core. An in-situ deposition method was used to produce uniform thin film coatings of the electronic polymer on the optical fiber. It was found that optimization ofthe sensor sensitivity can be achieved by selecting the proper incident wavelength, excitation conditions, and optical detection technique. Chemical sensors were developed and tested for detection ofHCl and NH3 vapors along with the reducing agent hydrazine. The results clearly demonstrate that conjugated polymer coated fiber optics represent a promising new approach for the detection of volatile toxic gasses.

Microacoustic devices have widely been investigated as a means of sensing mechanical, chemical, and electrical liquid properties. Shear-horizontally polarized modes might be used to avoid radiation losses into the liquid. If an acoustic waveguide is applied on top of the propagation surface, those modes can be converted into Love waves. Love mode devices can be designed to show one of the highest sensitivities among all microacoustic sensors while providing high mechanical robustness and being chemically inert. Good candidates for piezoelectric substrates are several Y-rotated quartz cuts as well as 36°YX-LiTaO3. SiO2 is a superior guiding layer because of low damping and excellent chemical and mechanical resistance. Besides high sensitivity, low temperature dependence is crucial in order to overcome the need of precision temperature control. This leads to the scope of the present work: Based on both experimental and theoretical results, the specific properties of various temperature-compensated YZ’-quartz/SiO2 and 36°YX-LiTaO3/SiO2 systems are discussed. For quartz based devices, temperature compensated systems could be realized with any sensitivity up to 80 % of the specific maximum, whereas the coupling coefficient being the limiting factor. In opposite, 36°YX-LiTaO3/SiO2 systems yield high coupling for any system of interest to sensor applications. However, there is only one temperature compensated configuration. This particular system provides a sensitivity of about 50% of the maximum of quartz based devices. In summary, for both material systems device configurations could be identified which combine high temperature stability, a suitable coupling coefficient, and high sensitivity.

Silicon sensors can be fabricated as small, rugged and reliable chip devices with a broad field of applications in medicine, biotechnology, food analysis and environmental monitoring. Thus, there is an increasing demand in realizing such sensors for the determination of, e.g. chemical and biological quantities in aqueous solutions. By developing semiconductor-based field-effect structures, moreover, their main advantage is due to the combination of both the physical effect as the transducer principle and the deposition of the sensitive layers directly onto the silicon chip. In this work, different sensor types that are originated from the field effect are presented: The capacitive ElS (electrolyte-insulator-semiconductor) sensor is suitable for the pH detection using the capacitance/voltage technique. By immobilizing an additional enzyme layer, e.g. of penicillinase, a biosensor has been realized. Both sensors can be integrated as an EIS sensor array. The utilization of the porous silicon technology offers the possibility of a further miniaturization. The LAPS (light-addressable potentiometric sensor) is based on the identical ElS structure. Here, each measuring point on the surface can be arbitrarily addressed by a probing light. The resulting photocurrent is generated as the sensor signal. This arrangement also allows a two-dimensional mapping of the spatial distribution of ions or molecules.

Identification of bacteria and other biological moieties finds a broad range of applications in the environmental, biomedical, agricultural, industrial, and military arenas. Linking these applications are biological markers such as fatty acids, whose mass spectral profiles can be used to characterize biological samples and to distinguish bacteria at the gram-type, genera, and even species level. Common methods of sample analysis require sample preparation that is both lengthy and labor intensive, especially for whole cell bacteria. The background technique relied on here utilizes chemical derivatization of fatty acids to the more volatile fatty acid methyl esters (FAMEs), which can be separated on a gas chromatograph column or input directly into a mass spectrometer. More recent publications demonstrate improved sample preparation time with in situ derivatization of whole bacterial samples using pyrolysis at the inlet; although much faster than traditional techniques, these systems still rely on bench-top analytical equipment and individual sample preparation. Development of a miniaturized pyrolysis/GC instrument by this group is intended to realize the benefits of FAME identification of bacteria and other biological samples while further facilitating sample handling and instrument portability. The technologies being fabricated and tested have the potential of achieving pyrolysis and FAME separation on a very small scale, with rapid detection time (1-10 min from introduction to result), and with a modular sample inlet. Performance results and sensor characterization will be presented for the first phase of instrument development, encompassing the microfabricated pyrolysis and gas chromatograph elements.

AT-cut quartz crystal plates vibrating in the thickness-shear mode are well known as mass sensitive devices, also called quartz crystal microbalances (QCM). After deposition of a sensitive layer on one or both surfaces of the quartz discs these resonators are suitable for the application as chemical sensors for analysis in gaseous and liquid media. Up to now the resonant frequencies of these resonators are 5 to 30 MHz. The application of combined photolithographic and etching processes offers new promising approaches for the manufacturing of quartz resonators with higher resonant frequencies, up to 250 MHz, and smaller diameters, leading to higher sensitivities of the sensor device. So called inverted mesa resonators were fabricated and subsequently characterized optically, mechanically, and electrically. The measured Q-factors are about 5 104 , which is excellent for high frequency resonators. The correlation between electrode diameter and inharmonic modes was investigated. The influences of surface roughness and etch channels on the resonators' performance were examined. The behaviour under acoustic load was investigated experimentally. In liquid media, changes in the viscosity (?L) and density (pL) lead to a decrease of the resonant frequency (?f) of the QCM. A linear relationship between (?LPL)12 and ?f was observed, in agreement with theory, while the frequency shifts are much higher than reported before.

We introduce a fiber optical sensor system that can be applied to environmental analysis. The compact system is based on laser-induced, time-resolved fluorescence emission spectroscopy. It uses a miniaturised all solid state laser which is operated at 266 nm as the excitation source and a spectrograph/image intensifier/CCD-camera for time-resolved detection of the fluorescence. The versatility of the instrument is demonstrated by the analysis of various substances using multivariate calibration techniques. Xylene could be analysed in natural water samples in the presence of other mono-aromatics with a detection limit of 10 ?g/l. The fluorescence tracer sulforhodamine G was analysed in river water with a detection limit of 10 ng/l. Furthermore the system is able to detect oil contaminations in soil in the concentration range of 100 ppm and above.

We report examples of the use of a scanning tunable CO2 laser lidar system in the 9-11 ?m region to construct images of vegetation and rocks at ranges of up to 5 km from the instrument. Range information is combined with horizontal and vertical distances to yield an image with three spatial dimensions simultaneous with the classification of target type. Reflectance spectra in this region are sufficiently distinct to discriminate between several tree species, between trees and scrub vegetation, and between natural and artificial targets. Limitations imposed by laser speckle noise are discussed.

A compact, tunable and single-mode laser for the mid-infrared (MIW) spectral range is developed by difference frequency generation (DFG) in AgGaS2 and two "off-the-shelf' diode lasers. The MIR laser light is coupled into a silverhalide fiber and at the end of the fiber an infrared detector is used to record the transmitted MIR laser light. If the index of refraction of the fiber material is higher than the one of the surrounding medium the light is guided through the fiber due to total reflection. There are two loss mechanisms that will attenuate the laser intensity when passing through the fiber: (1) The frustrated total reflection (FTR) and (2) the attenuated total reflection (ATR). The FTR is related to a change of the index of refraction while the ATR is related to a change of the absorption coefficient. When tuning the MIR laser over an absorption line of a molecule that is outside of the fiber both, FTR and ATR, contribute to the measured spectral line profile that is recorded by the infrared detector. Similar to the direct laser absorption spectroscopy the recorded line profiles in the case of the evanescent~field spectroscopy can be used for estimating concentrations of molecules. A practical application of the evanescent field fiber sensor is shown as H2S is measured online and insitu at the volcano "Solfatara" in Italy.

Conductometric gas microsensors offer the benefits of ppm-level sensitivity, real-time data, simple interfacing to electronics hardware, and low power consumption. The type of device we have been exploring consists of a sensor film deposited on a "microhotplate"- a 100 micron platform with built-in heating (to activate reactions on the sensing surface) and thermometry. We have been using combinatorial studies of 36-element arrays to characterize the relationship between sensor film composition, operating temperature, and response, as measured by the device’s sensitivity and selectivity. Gases that have been tested on these arrays include methanol, ethanol, dichloromethane, propane, methane, acetone, benzene, hydrogen, and carbon monoxide, and are of interest in the management of environmental waste sites. These experiments compare tin oxide films modified by catalyst overlayers, and ultrathin metal seed layers. The seed layers are used as part of a chemical vapor deposition process that uses each array element’s microheater to activate the deposition of SnO2, and control its microstructure. Low coverage (2 nm) catalytic metals (Pd, Cu, Cr, In, Au) are deposited on the oxides by masked evaporation or sputtering. This presentation demonstrates the value of an array-based approach for developing film processing methods, measuring performance characteristics, and establishing reproducibility. It also illustrates how temperature-dependent response data for varied metal/oxide compositions can be used to tailor a microsensor array for a given application.

We have developed a design for a neutral wind instrument for the upper atmosphere based on arrays of miniature pressure sensors. The elements of the array are tunnel displacement transducers (TDT's). We have characterized the TDT, a microelectomechanical system (MEMS), and evaluated TDT capabilities for the determination of pressure within a 2—D array. This is an approach for direct measurement of neutral winds. The sensitivity of the TDT is critical for this application. The output signal of a TDT is the bias voltage dependent tunneling current. This is a differential signal in a force balanced TDT due to the fact that a restoring force is applied to the diaphragm in order to restore the tunneling current to a fixed value. We have fabricated and assembled a custom test apparatus for the TDT based sensor arrays. We have demonstrated simultaneous collection of data from 4 TDT's that were hybrid assemblies of monolithic 2-TDT arrays. Results showed that correlation of the tunneling signal to the actual bias voltage difference required to balance the displacement force gives reasonable information for a pressure sensor and defines the required signal processing.

Intelligent signal processing algorithms are shown to improve identification rates significantly in chemical sensor arrays. This paper focuses on the use of independently derived sensor status information to modify the processing of sensor array data by using a fast, easily-implemented "best-match" approach to filling in missing sensor data. Most fault conditions of interest (e.g., stuck high, stuck low, sudden jumps, excess noise, etc.) can be detected relatively simply by adjunct data processing, or by on-board circuitry. The objective then is to devise, implement, and test methods for using this information to improve the identification rates in the presence of faulted sensors. In one typical example studied, utilizing separately derived, a-priori knowledge about the health of the sensors in the array improved the chemical identification rate by an artificial neural network from below 10 percent correct to over 99 percent correct. While this study focuses experimentally on chemical sensor arrays, the results are readily extensible to other types of sensor platforms.

In this paper, a digital imaging system was used to analyze a series of colorimetric reagent dots. A commercially available digital camera was used with in-house developed software to capture and analyze data. The color captured was digitized and broken down into three 8-bit color values (red, green and blue). The data was then processed to generate distinct recognizable patterns that represent the original color input. A pH indicator sensor array was used to demonstrate the practicality of the system. Preliminary results show that the system is able to discriminate samples over a wide pH range (pH 1-13) and resolve samples to within ± 0.5 pH units.

An approach for coordinated, distributed, in situ sensing is proposed. An example application is chosen (offshore marine monitoring) and the approach is described in the context of the example. Three major research challenges are discussed, (1) coordinated sensing and triggering, (2) localization and time synchronization, and (3) micro mobility.

A new algorithm for the design of optical computing filters for chemical analysis otherwise known as Multivariate Optical Elements (MOEs), is described. The approach is based on the nonlinear correlation of the MOE layer thicknesses to the standard error in sample prediction for the chemical species of interest using a modified version ofthe Gauss-Newton nonlinear optimization algorithm. The design algorithm can either be initialized by random layer thicknesses or by a pre-existing design. The algorithm has been successfully tested by using it to design a MOE for the determination of copper uroporphynn in a quaternary mixture of uroporphyrin (freebase), nickel uroporphyrin, copper uroporphynn, and tin uroporphyrin.

A significant phase in the development of an intelligent agent is the construction of its Knowledge Base (KB) on the basis of which it has to take the appropriate actions. The validation and verification (V&V) of KBs is an important part of any KB system development, ignoring it can result anomalies during run-time. The paper discusses the implementation of a utility for validation and verification of KBs'. The methodology transforms the rules in a KB to an equivalent Petri net representation and then applies the analytical tools ofthe Petri net theory for the detection of errors.

Various aspects of controlling the environment can be structured in a form of multiple-criteria decision making. The combination ofniathematical modeling, decisioninaking, and the use ofpattern recognition in the environmental modeling and management of complex systems is discussed from a historical perspective. Of particular interest is the use of objective space technology in modeling multiple criteria decisionmaking. This paper presents a model and solution algorithm for such type ofdecision problem using a multiple objective linear program. Several applications for environmental control can be modeled using this methodology.

Spectral pattern recognition (SPR) methods are among the most powerful tools currently available for noriinvasively examhiin the spectroscopic and other chemical data for environmental analysis and monitoring. Using spectral data, these systems have found a variety of applications in chemometric systems such as gas chromatography, fluorescence spectroscopy, etc. An advantage of SPR approaches is that they make no a priori assumption regarding the structure of spectra. However, a majority o these systems rely on humanjudgment for parameter selection and classification. We considered a SPR problem as a composite of five subproblems: pattern acquisition, feature extraction, feature selection, knowledge organization, and pattern classification. One ofthe basic issues in SPR approaches is to determine and measure the features useful for successful classification. Selection of features that contain the most discriminatory information is important because the cost of pattern classification is directly related to the number offeatures used for classification. Various features present in a pattern and a large variety of classification algorithms could be used. A spectral pattern classification system combining the above components and multivariate decisiontheoretic approaches for classification is developed. It is shown how such a system can be used for large data analysis, warehousing, and interpretation. In a preliminary test, the system was used to classif' synchronous UV-vis fluorescence spectra ofrelatively similar petroleum oils with reasonable success.

The broader definition of chemometrics includes methods such as pattern recognition (PR) and signal/image processing for noninvasive analysis and interpretation of data. These methods are among the most powerful tools currently available for noninvasively examining spectroscopic and other chemical data. Using spectral data, these systems have found a variety of applications employing analytical techniques for gas chromatography, fluorescence IR or NMR spectroscopy, etc. An advantage of PR approaches is that they make no a priori assuniption regarding the stmcture of the spectra. However, a majority of these systems rely on hunianjudgment for parameter selection and classification of spectra. Generally a spectral pattern recognition (SPR) problem is considered as a group of several subproblems. We considered a SPR problem as a group of five subproblems: spectra acquisition, feature extraction, feature selection, spectra organization, and spectra classification. One of the basic issues in PR approaches is to determine and measure the discriminatory features useful for successful classification. A spectral pattern classification system, combining spectral feature extraction and selection, and decision-theoretic approaches, is developed. It is shown how such a system can be used for analysis of large data analysis, warehousing, and interpretation.

Among other advantages over their traditional, full-size counterparts, micro-plasmas present the possibility of cost savings due to a reduction in the flow-rate of expensive plasma support gases such as argon or helium. Despite their advantages, micro-plasmas are not often used because they cannot handle the relatively large sample loads produced by the traditional pneumatic nebulization sample introduction systems typically used in full-size plasmas when analyzing liquid samples. This paper describes steps taken to couple an in-torch vaporization (ITV) "dry" sample introduction system to a dc micro-plasma (which has been designed around a micro-fluidics micro-channel) for use in environmental analysis of liquid micro-samples by atomic emission spectrometry (AES).