D. D. Markushev

Low-cost, portable photoacoustic setup for solid samples

We have developed a low-cost, portable photoacoustic instrument. The device consists of a detection unit comprising a photoacoustic cell with an embedded laser diode or a light-emitting diode, a photodiode, an electret microphone (60 ? 40 ? 40 mm3), and a signal processing and power supply unit in a box containing batteries and electronics (160 ? 140 ? 60 mm3). A PC or portable computer is required to operate the device and for data processing. The weight of the instrument without the computer is 1.70 kg. The computer, or more precisely, its sound card is the essential part of the apparatus because it generates the signal for the laser diode or light-emitting diode modulation and processes signals from the microphone and photodiode. The computer sound card is used as a dual-phase lock-in amplifier. The software used for the control of the setup was also developed in course of this work. The photoacoustic instrument presented here allows measurements and quantitative analysis of numerous solid-state samples. It is simple in design and use, having a reasonable weight and portability.

Pulsed photoacoustic system calibration for highly excited molecules

The proposal of a simple method for pulsed photoacoustic system calibration is presented. Analysis of the photoacoustic signal shape is performed with different types of absorbing molecules (SF6 and C2H4) and the same buffer gas (Ar). Different total pressures (ptotal) of such gas mixtures and different laser fluences are used to obtain experimental results and compare them with theory. Results from such a comparison are used directly for the calculation of a calibration curve. A single experimental point calibration procedure for the case of highly vibrationally excited absorbing SF6 molecules is confirmed.

Pulsed photoacoustic system calibration for highly excited molecules: II. Influence of the laser beam profile and the excitation energy decay

Previously, we developed a simple semi-theoretical calibration procedure for a pulsed photoacoustic set-up. Also, we analysed the simplest, idealized case: top hat spatial profile of the laser beam and an exponential decay of excitation energy. The spatial profile of the laser beam is usually considered to be top hat or Gaussian in the photoacoustic measurements. In reality, there are always small discrepancies. Also, in these measurements the excitation energy decay is usually considered to be an exponential one. This assumption is commonly valid. Still, a non-exponential decay can exist as well. In this paper, we have examined the influence of these discrepancies on the measurement of the vibrational-to-translational relaxation time, as well as on the calibration of the photoacoustic set-up. We have theoretically examined gas mixtures in the case of pulsed excitation (multiphoton regime). Then, we have verified theoretical conclusions in one experimental example. It has been shown that the non-ideal profile and excitation energy decay can significantly influence the measurement of the vibrational-to-translational relaxation time. Also, it has been shown that they do not significantly influence the calibration of the photoacoustic set-up.

Effect of the absorption coefficient of aluminium plates on their thermoelastic bending in photoacoustic experiments

The open-cell photoacoustic signal measured in the transmission configuration for aluminum thin plates with thicknesses of 280 μm, 197 μm, and 112 μm is experimentally and theoretically analyzed, in the 20 Hz–7 kHz modulation frequency range. It is shown that the observed differences between the predictions of the standard thermoelastic model and the experiment data of both the amplitude and phase of the photoacoustic signal can be overcome by considering the aluminum samples coated with a thin layer of black paint as volume-absorber materials. This new approach provides a quite good agreement with the obtained experimental data, in the whole frequency range, and yields an effective absorption coefficient of (16 ± 2) mm−1, for a 280 μm-thick sample. The introduction of the finite absorption coefficient led to the correct ratio between the thermal diffusion and thermoelastic components of the photoacoustic signal. Furthermore, it is found that the “volume-absorber” approach accurately describes the behavior of the amplitude, but not that of the phase recorded for a 112 μm-thick sample, due to its relatively strong thermoelastic bending, which is not considered by this theory. Within the approximation of the small bending, the proposed “volume-absorber” model provides a reliable description of the photoacoustic signal for Al samples thicker than 112 μm, and extends the applicability of the classical “opaque” approach.

Laser beam spatial profile determination by pulsed photoacoustics: exact solution

The pulsed photoacoustic spectroscopy technique is one of the oldest but respectable techniques which offer some unique features that are relevant to trace gas monitoring. Despite obvious advantages there are some drawbacks that have so far alienated this spectroscopy technique from routine air monitoring. One of them is the additional instrumentation necessity to fulfill the requirement for an accurate knowledge of the spatial and temporal profiles of the excitation light source. Here we present one method for a laser beam spatial profile determination and simultaneous vibrational–translational relaxation time calculation. It is based on the temporal shape analysis of the photoacoustic signals. Symmetric laser beam spatial profiles are obtained with a high level of accuracy. Even though this method does not work in real time, it gives a good basis for future beam profile investigations. Experimental signals used in our analysis are obtained after infrared multiphoton absorption in the SF6–Ar mixture.

Photoacoustic elastic bending in thin film–substrate system: Experimental determination of the thin film parameters

The aim of this work is the development of photoacoustic (PA) method for the measurement and determination of parametres of thin films (with a thickness of less than 1 μm). Experimental study of the optical, thermal, and elastic characteristics of the thin film on Si substrate by PA elastic bending method was given. Thin film–semiconductor (Si) sample is modeled by simultaneous analysis of the plasma, thermal, and elastic wave equations. Two normalization procedures of the PA elastic bending signal in function of the modulation frequency of the optical excitation were established. The experimental PA elastic bending signals were measured and analysed. Without loss of generality, the TiO2 thin film (with a thickness of 0.5 μm) on Si substrate (circular plate) was experimentaly studied. We have studied the PA elastic bending signals in order to obtain the values of optical, thermal, and elastic parameters of TiO2 film. The analysis shows that it is possible to develop noncontact and nondestructive experimental method—PA elastic bending method for thin film study, with possibility to obtain the optical, thermal, and elastic parameters of the film thinner than 1 μm.

Photoacoustic elastic bending in thin film—Substrate system

Theoretical model for optically excited two-layer elastic plate, which includes plasmaelastic, thermoelastic, and thermodiffusion mechanisms, is given in order to study the dependence of the photoacoustic (PA) elastic bending signal on the optical, thermal, and elastic properties of thin film—substrate system. Thin film-semiconductor sample (in our case Silicon) is modeled by simultaneous analysis of the plasma, thermal, and elastic wave equations. Multireflection effects in thin film are included in theoretical model and analyzed. Relations for the amplitude and phase of electronic and thermal elastic bending in the optically excited two-layer mechanically-supported circular plate are derived. Theoretical analysis of the thermodiffusion, plasmaelastic, and thermoelastic effects in a sample-gas-microphone photoacoustic detection configuration is given. Two normalization procedures of the photoacoustic elastic bending signal in function of the modulation frequency of the optical excitation are established....

Photoacoustic signal and noise analysis for Si thin plate: Signal correction in frequency domain

Methods for photoacoustic signal measurement, rectification, and analysis for 85 μm thin Si samples in the 20-20 000 Hz modulation frequency range are presented. Methods for frequency-dependent amplitude and phase signal rectification in the presence of coherent and incoherent noise as well as distortion due to microphone characteristics are presented. Signal correction is accomplished using inverse system response functions deduced by comparing real to ideal signals for a sample with well-known bulk parameters and dimensions. The system response is a piece-wise construction, each component being due to a particular effect of the measurement system. Heat transfer and elastic effects are modeled using standard Rosencweig-Gersho and elastic-bending theories. Thermal diffusion, thermoelastic, and plasmaelastic signal components are calculated and compared to measurements. The differences between theory and experiment are used to detect and correct signal distortion and to determine detector and sound-card characteristics. Corrected signal analysis is found to faithfully reflect known sample parameters.

Photoacoustic elastic bending method: Study of the silicon membranes

Photoacoustic (PA) amplitude and phase spectra vs the modulation frequency of the excitation optical beam were measured and analyzed by PA elastic bending method. The PA spectra were measured in a frequency range (from 20 to 20000 Hz), for different thicknesses of Si rectangular membranes (from 10 to 100 μm).. The elastic characteristics of the Si chip with square membranes were described by the theory of thin elastic plate, i.e. as an elastic simply supported rectangular plate. The PA signals (sum of the thermodiffusion, thermoelastic and electronic deformation components) were calculated and compared with the experimental ones. These results showed that the PA elastic bending spectra are convenient for investigation the characteristics of micromechanical membraines. The PA elastic bending spectra of the optically driven micro-opto-electro-mechanical systems (MOEMS) enable, for example, to investigate the different electronic and thermal transport characteristics or technological processes in MEMS fabrications, etc.

Computationally intelligent pulsed photoacoustics

In this paper, the application of computational intelligence in pulsed photoacoustics is discussed. Feedforward multilayer perception networks are applied for real-time simultaneous determination of the laser beam spatial profile and vibrational-to-translational relaxation time of the polyatomic molecules in gases. Networks are trained and tested with theoretical data adjusted for a given experimental set-up. Genetic optimization has been used for calculation of the same parameters, fitting the photoacoustic signals with a different number of generations. Observed benefits from the application of computational intelligence in pulsed photoacoustics and advantages over previously developed methods are discussed, such as real-time operation, high precision and the possibility of finding solutions in a wide range of parameters, similar to in experimental conditions. In addition, the applicability for practical uses, such as the real-time in situ measurements of atmospheric pollutants, along with possible further developments of obtained results, is argued.