The use of in-situ chemical sensors in semiconductor wafer fabrication will increase significantly over the next five years. Advanced gas phase and liquid phase real time, non contaminating chemical sensors will be applied in four major categories of use. These four areas of application are: (1) Upstream measurements of trace contaminants in high purity process gases and liquids. (2) In-the-chamber or in-the- process bath measurements of chemical species of on line active process tools. (3) Real time, on line, downstream measurements of gas phase and liquid phase effluents coming from semiconductor wafer process tools. (4) Semiconductor industrial hygiene and safety monitors for toxic substances in the work place. This paper will review each of these major applications for both gases and liquids as they will need to be integrated into semiconductor wafer manufacturing over time. Some existing sensor applications currently exist. This review will focus extensively on in-situ real time chemical monitoring sensors.

A Modeling and Simulation study of the limits of remote detection by passive IR has led to a new concept for the remote detection of hazardous clouds. A passive IR signature model was developed using as input the ERDEC IR spectral data bases for chemicals and biologicals, and the atmospheric transmittance model MODTRAN (MODerate resolution TRANSmittance). The cloud travel and dispersion model VLSTRACK (Vapor, Liquid, and Solid TRACKing) was used to simulate chemical and biological clouds. An easily applied spectral discrimination technique was developed using Linear Programming. All of these were melded with MATHEMATICA to produce images of 3 threat clouds: Sarin, mustard and an unnamed biological. (For reasons of space only Sarin is discussed here.) The HAZCI (HAZarous Cloud Imager) is a unique configuration of a spatially scanning Fourier Transform IR on the same level of complexity as the M21, but capable of orders-of-magnitude greater sensitivity and moving operation. The concept provides for automatic detection and operator assisted discrimination for chemical clouds and biological clouds at ranges greater than 50 kms. Both chemical and biological clouds can generally be discriminated from other known battlefield contaminants and from all materials (including biologicals) that are uniformly distributed in the atmosphere; but, specific detection of pathogenic biologicals is not projected at this time.

Many chemical warfare agents are dispersed as small aerosol particles. In the past, most electro-optical excitation and detection schemes have used continuous or pulsed lasers with pulse lengths ranging from nanoseconds to microseconds. In this paper, we present interesting ongoing new results on femtosecond imaging and on the time dependent solutions to the scattering problem of a femtosecond laser pulse interacting with a single small aerosol particle. Results are presented for various incident pulse lengths. Experimental imaging results using femtosecond pulses indicate that the diffraction rings present when using nanosecond laser pulses for imaging are greatly reduced when femtosecond laser pulses are used. Results are presented in terms of the internal fields as a function of time and the optical size parameter.

Photothermal detection methods use the light-absorbing sample material as the sensor transducer. Photothermal displacement methods using interferometric detection will be used to measure environmental analytes particularly if those analytes can be incorporated into the sensor probe. We are using various nanoporous optical sol-gel materials in sensor probes for photothermal detection. Photothermal sensors made using sol-gel materials have advantages for infrared and near infrared photothermal in situ detection including improved sensitivity, lower limits of detection, easier miniaturization, and room-temperature operation. The potential for photothermal interferometric near infrared detection of organophosphorous compounds adsorbed in silica sol-gel is demonstrated.

This paper quantifies the temperature and humidity dependence of a polymer-based gas sensor. The measurement and analysis of three polymers indicates that resistance changes in the polymer films, due to temperature and humidity, can be positive or negative. The temperature sensitivity ranged from +1600 to -320 ppm/ degree(s)C and the relative humidity sensitivity ranged from +1100 to -260 ppm/%. These results were obtained from three-day experiments in an air ambient where the gas sensors were kept in stagnant air to allow for equilibration of the gas sensors and instruments used to measure the temperature and humidity. In spite of significant variations in temperature and humidity, the detection of 50 ppm resistance changes is possible.

The brief description of new lidar prototype for remote chemical monitoring and profiling in the 8 - 12 micron range is given. The lidar includes a Nd:YAG laser (1 J per pulse) source, optical parametrical oscillator (0.2 J per pulse), and four-wave Raman emitter (20 mJ output per pulse). The receiver consists of the hydrogen SRS cells, pumped by an additional OPO pulse. Sensitivity of this receiver reaches approximately 1000 photons per pixel. The applications of this lidar for remote detection of chemicals in atmosphere will also be discussed.

Advances in the development of crystals with good nonlinear properties have made optical parametric oscillators (OPOs) strong candidates for generation of coherent radiation. These technological advancements have renewed research in the development of solid state 8 - 12 micrometers coherent OPO LIDAR sources for remote chemical sensing applications, if millijoule pulse energies are adequate and short (ns) pulses are beneficial. We present recent advances in OPO technology that have generated tunable, 3 - 5 micrometers and 8 - 12 micrometers radiation. Specific pump source technologies to be addressed are CO₂, Nd:YAG, Er and Ho:host materials. The paper will examine nonlinear materials such as ZnGeP₂, AgGaSe₂, AgGaS₂, and CdSe and their relevant parameters.

First steps in the development of a new method for the remote detection of gaseous emanations by FTIR are presented. The method is based on the use of a double beam interferometer-spectrometer (CATSI) optimized for optical subtraction. The instrument is described with a particular emphasis on its capabilities for differential detection and background suppression. Radiometric equations applicable to the differential detection of gaseous emanations by FTIR are established for the general case of slant path scenarios containing any type of background scenes. Finally the potential of the method is illustrated through several laboratory measurements done with the CATSI system.

A novel scheme of a laser-based chemical sensor has been examined. The scheme is based on the lasing frequency shift of a DBR laser as a result of refractive index change of the sensitive coating in the presence of chemicals in question. The applicability and advantages of different schemes are discussed. The results of preliminary experiments related to the construction and stability of an external cavity DBR laser and interferometric measurements of refractive index change are presented.

Recently, Bomem developed CATSI, a small FTIR system referred to as the Compact ATmospheric Sounding Interferometer. In order to meet the highest radiometric precision and accuracy, the instruments performs calibrations, using observations of hot and cold blackbody reference sources as the basis for a two-point calibration at each wavenumber. This paper presents the results of the analysis of the radiometric calibration of this instrument, with emphasis on the temporal behavior of the instrument response.

A FT-IR spectrometer thermal stability, responsivity and self emission was evaluated under controlled laboratory conditions. Internal diagnostics provided by the FT-IR spectrometer design insured the acquisition of accurate and precise spectral data. Absorbance measurements performed using the spectrometer agreed to within 3% of the literature values. The internal polystyrene film permitted a reliable assessment of the wavenumber axis registration which is critical in the identification of target analyte spectral features.

Raman spectroscopy has been shown to be a useful tool for characterizing neat crystalline explosive samples and for identifying principle components in many propellant and explosive formulations. Recently, we have been investigating changes in Raman spectra of explosives and propellant formulations which occur as the temperature approaches the melting point of the sample. We report recent measurements of Raman spectra of explosives and propellant formulations during bulk heating, and recent measurements of laser heating of the samples during measurement of Raman spectra. The results of these measurements are important to investigators using Raman spectroscopy to measure vibrational spectra at the surface of burning propellant samples.

Dwindling research and development funds coupled with increasing customer demands and rapidly developing technologies present a difficult dilemma to systems developers. A current area of innovation is the integration of the abundant but diverse computer resources (both hardware and software) into practical and powerful computing environments to support the product development processes in the industrial and defense sectors. Edgewood has leveraged a number of existing resources to create a `virtual prototyping' environment based on the principles of Concurrent Multi-Level Simulation. We named our environment the `Computational Prototyping Network'.

A new type of imaging spectrometer is being developed for remote sensing and chemical detection. It is based on a rotary Fourier Transform Spectrometer design. The rotary nature of the scan allows both high speed operation and laserless sampling of the detector signal. Its small size also allows cryogenic operation for enhanced detection capability and stable calibration. A 360 scan per single pixel prototype was built and demonstrated at ERDEC in April 1996. The single scan time at this speed is just under 1 millisecond. The optical sensor measures 6' X 4' X 2', and weighs approximately 3 lbs. Spectral resolution is 2 wavenumbers (cm^-1) maximum, and coverage is 12 - 14 microns using Zinc Selenide (ZnSe) optics. Power consumption is less than 1 watt at steady state, slightly more at spinup. Electronics and software package used COTS 12 bit, 10 MHz data acquisition and DSP board in conjunction with Labview software for acquisition and analysis on a desktop PC. Following this demonstration, the scan linearity was studied by passing a 3.3 micron laser signal through the IR channel using different rotor materials, and using laserless sampling to look at the resulting lineshape. The results of that testing is the subject of this paper. Subsequently, a 3 by 3 HgCdTe (MCT) detector array is being mated to the optical sensor. An enhanced data channel with 16 simultaneous 12 bit inputs is also being fitted, running at 1/10 of original speed. Image quality and pixel crosstalk are being analyzed at this lower speed. Field measurements will be made in summer 1997.

A methodology is described for an airborne, downlooking, longwave infrared imaging spectrometer based technique for the detection and tracking of plumes of toxic gases. Plumes can be observed in emission or absorption, depending on the thermal contrast between the vapor and the background terrain. While the sensor is currently undergoing laboratory calibration and characterization, a radiative exchange phenomenology model has been developed to predict sensor response and to facilitate the sensor design. An inverse problem model has also been developed to obtain plume parameters based on sensor measurements. These models, the sensors, and ongoing activities are described.

Frequency-agile lidar (FAL) with the capability of tuning to more than 60 wavelengths within 1 second, can provide better vapor detection and multimaterial discrimination ability than two-wavelength differential absorption lidar. This paper extends an earlier optimal approach for processing FAL data to include signals with a fluctuating component due to shot-to-shot variations in the transmitted pulse energy whenever a local measurement of that energy is available. Traditional methods for performing this so-called transmit energy normalization have typically ratioed the received lidar signal by the energy monitor data. It is shown that this is, in general, not only a suboptimal approach, but can degrade the performance of the lidar below that achieved by not normalizing the data at all. The optimal approach is shown to be a linear correction to the received signal proportional to the monitor data. The estimated correlation between the transmitted and received signals provides the optimal proportionality factor at each wavelength. Simple approximate expressions are derived for comparing the performance of the optimal versus ratio estimators. The approach is illustrated on both synthetic and actual FAL data.

A new model for generating synthetic images of plumes has been developed using a radiometrically based ray-tracing algorithm. Existing plume models that describe the characteristics of the plume (constituents, concentration, particulate sizing, and temperature) are used to generate AutoCAD models for input into the code. The effects of scattered light using Mie theory and radiative transfer, as well as thermal self-emission/absorption from within the plume, are modeled for different regions of the plume. The ray-tracing accounts for direct sunlight, scattered skylight, reflected sunlight from the background, and thermal self-emission from both the atmosphere and background. Synthetic generated images of a cooling tower plume, composed of water droplets, and a factor stack plume, composed of methyl chloride, are produced for visible, MWIR, and LWIR bands. Images of the plume from different view angles are also produced. Observations are made on the interaction between the plume and its background and possible effects for remote sensing. Images are made of the methyl chloride plume in which the concentration and temperature are varied to determine the sensitivity of the radiance reaching the sensor.

The Night Vision and Electronic Sensors Directorate's (NVESD's) Signatures and Sensors Branch performed a series of experiments to evaluate the feasibility of developing a man portable LIDAR system to perform column content and/or range resolved chemical detection. In order to perform these experiments in an expedient and low cost manner an existing Frequency Agile Laser system was utilized. These experiments were conducted in three phases: (1) full system characterization, (2) chemical cell tests, and (3) simulated stack plume tests. The system characterization was used as a baseline for system performance. Once the system was fully characterized, a series of chemical cell experiments were conducted. The cell test determined to basic feasibility of column content chemical detection and quantification using LIDAR technology. Following a series of successful chemical cell experiment, the stack tests provided a more extensive and realistic way of evaluating column content mode LIDAR capabilities as well as the ability to evaluate the feasibility of performing range resolved chemical detection. Four chemicals were used for the cell and stack experiments. The chemicals were chosen based on real chemical stack exhaust of production plant processes. This paper provides an overview of the LIDAR experiments performed, preliminary data results and analysis, and future efforts planned in support of this project.