A Chemiluminescent Analyzer for Stratospheric Measurements of the Ozone Concentration (FOZAN)

V. Yushkov Central Aerological Observatory, Dolgoprudny, Russia

Search for other papers by V. Yushkov in
Current site
Google Scholar
PubMed
Close
,
A. Oulanovsky Central Aerological Observatory, Dolgoprudny, Russia

Search for other papers by A. Oulanovsky in
Current site
Google Scholar
PubMed
Close
,
N. Lechenuk International Sakharov Institute of Radioecology, Minsk, Belarus

Search for other papers by N. Lechenuk in
Current site
Google Scholar
PubMed
Close
,
I. Roudakov International Sakharov Institute of Radioecology, Minsk, Belarus

Search for other papers by I. Roudakov in
Current site
Google Scholar
PubMed
Close
,
K. Arshinov Institute of Technical Acoustic, Vitebsk, Belarus

Search for other papers by K. Arshinov in
Current site
Google Scholar
PubMed
Close
,
F. Tikhonov Institute of Technical Acoustic, Vitebsk, Belarus

Search for other papers by F. Tikhonov in
Current site
Google Scholar
PubMed
Close
,
L. Stefanutti IROE–CNR, Florence, Italy

Search for other papers by L. Stefanutti in
Current site
Google Scholar
PubMed
Close
,
F. Ravegnani FISBAT–CNR, Bologna, Italy

Search for other papers by F. Ravegnani in
Current site
Google Scholar
PubMed
Close
,
U. Bonafé FISBAT–CNR, Bologna, Italy

Search for other papers by U. Bonafé in
Current site
Google Scholar
PubMed
Close
, and
T. Georgiadis FISBAT–CNR, Bologna, Italy

Search for other papers by T. Georgiadis in
Current site
Google Scholar
PubMed
Close
Full access

We are aware of a technical issue preventing figures and tables from showing in some newly published articles in the full-text HTML view.
While we are resolving the problem, please use the online PDF version of these articles to view figures and tables.

Abstract

Within the framework of the Airborne Polar Experiment, a new ozone analyzer based on chemiluminescent technology has been developed for high-altitude measurements. The instrument was tested on board the M-55 Geophysica aircraft during some flights over Italy, reaching an altitude of about 20 000 m and a temperature of about −75°C. This paper presents the chemiluminescent characteristics of the sensor, the electronic design of the instrument, and a comparison between the results obtained during a test flight with those derived from a contemporary balloon sounding. The instrumental performances of the analyzer were found to be suitable for stratospheric applications.

Corresponding author address: Dr. Teodoro Georgiadis, CNR National Research Council, FISBAT Institute, Via Gobetti 101, I-40129 Bologna, Italy.

Abstract

Within the framework of the Airborne Polar Experiment, a new ozone analyzer based on chemiluminescent technology has been developed for high-altitude measurements. The instrument was tested on board the M-55 Geophysica aircraft during some flights over Italy, reaching an altitude of about 20 000 m and a temperature of about −75°C. This paper presents the chemiluminescent characteristics of the sensor, the electronic design of the instrument, and a comparison between the results obtained during a test flight with those derived from a contemporary balloon sounding. The instrumental performances of the analyzer were found to be suitable for stratospheric applications.

Corresponding author address: Dr. Teodoro Georgiadis, CNR National Research Council, FISBAT Institute, Via Gobetti 101, I-40129 Bologna, Italy.

1. Introduction

Since 1989 coordinated research has been planned in Europe with the aim of achieving a better understanding of global ozone depletion mechanisms. The effort in field and laboratory measurements, as well as atmospheric modeling, was particularly strong during the realization of the European Arctic Stratospheric Ozone Experiment (EASOE, 1991–92) and the Second European Stratospheric Artic and Midlatitudes Experiment (SESAME, 1994–95) campaigns. The results obtained demonstrated clearly that the chemical depletion of ozone occurred also within the Arctic vortex (WMO 1995).

At the end of November 1993, during the second Airborne Polar Experiment (APE) workshop, it was stressed how, with a heavy payload on board, the M-55 Geophysica stratospheric platform was able to reach regions sited in the eastern part of the polar vortex never previously investigated with any aircraft (Stefanutti et al. 1995).

Remote sensing instruments based on satellite platforms can monitor such regions (for a complete review see Kramer 1996), but they cannot substitute for in situ measurements, which are more precise and suitable for management in specific experiments, where the time, duration, path, and “geometry” of the mission can be chosen.

The Italian Antarctic Program (PNRA) submitted to the Russian Aviaecocentre a feasibility study for the installation of scientific instruments on board the M-55;it also partly funded the development of an in situ ozone analyzer prototype, jointly managed by Russian, Belarussian, and Italian scientific institutions, to be installed on board the M-55 aircraft.

Chemiluminescence sensors are most appropriate for use in stratospheric applications, specifically in the estimation of ozone and nitrogen oxide concentrations, thanks to their high sensitivity and very fast response time. The methodology is based on the heterophase selective chemiluminescence of a solid matrix filled with a mixture of dye and gallic acid produced by reaction with some atmospheric trace gases.

Laboratory equipment based on chemiluminescence for ozone determination have been described by Hodgeson et al. (1970) and Sahand et al. (1986); results of field tests have been presented by Ray et al. (1986), Güsten et al. (1992), and Bersis and Vassilion (1996).

This paper presents the results of laboratory experiments on a new ozone analyzer that employs a chemiluminescent sensor alongside the measurements performed during a test flight on board the Geophysica platform.

2. Sensor characteristics

Both the experimental results and the analyzer’s metrological characteristics are mainly determined by those of the chemiluminescent sensor, which is based on the heterophase luminescence of ozone in a solid-state matrix filled with gallic acid dye. The sensor characteristics were thoroughly investigated by means of a special experimental setup, consisting of a high-frequency ozone generator, a pneumatic pump, a reaction vessel (with the sensor placed inside it), and a chemiluminescence registering system. The ozone generator made it possible to obtain calibrated ozone concentration values within a 5–1400-ppb range at atmospheric pressure.

A small-sized membrane-type pump circulated air through the system at a rate of 0.2 L m−1. Gas flow containing ozone blew over the sensible surface of the sensor and was removed from the vessel through a central hole in a glass shield placed between the luminescent surface and photomultiplier.

The recording system comprised a photomultiplier (FEU-110), an amplifier, and a self-recorder or oscillograph.

The sensors were manufactured with backings made of filter papers, porous glasses, porous polymers, magnesium oxides plates, and fine-porous borosilicate glasses or thin-layer chromatographic plates with silica gel.

The solutions utilized to activate the sensible surface of the sensor were composed of gallic acid (pure for analysis) and chromatographically purified dyes such as rodamin B, coumarin 47, coumarin 102, eosine “H,” and POPOP. Acetone, diethyl ether, and ch.d.a.-type ethanol were used as solvents since they are readily soluble in dyes and gallic acid.

After being impregnated with solution, the sensors were dried in a vacuum at 60°C. Subsequently, the sensors underwent activation by means of a constant airflow of 1100-ppb ozone concentration, at atmospheric pressure, while chemiluminescent intensity was simultaneously measured.

During the activation process, the chemiluminescent intensity of most compositions increases approximately linearly with time, follows and reaches a maximum, and then remains constant. Activation time is determined by the type of dye adopted, and for sensors using coumarin 30 and rodamin B, was about 100 and 180 min, respectively.

From the activation curve in Fig. 1, for sensors with coumarin 47, it can be seen how, having reached its maximum, the sensitivity decreases, oscillating for some 15 min before becoming stable.

The sensors’ operational stability is largely determined by the type of backing used: for course-grained backings—that is, porous polymer substances, glass, and paper filters—the scatter of signals under stable operational conditions reaches 30%, while for fine-grained borosilicate glass (MgO thin-layer chromatographic plates) it is no higher than 5%.

Figure 2 presents the kinetic characteristics of sensors (coumarin 47–and rodamin B–based) using dye chromatographic plates. The periods of signal growth or reduction are dependent on the gas circulation rate. The investigations conducted showed that the most acceptable performances, in terms of sensitivity and operational stability, was displayed by coumarin 47–based sensors. At increased ozone concentrations the sensitivity of sensors using rodamin B and coumarin 30 rose during the first instants of time to exceed its stationary value. In addition, sensor operational stability decreased.

Sensitivity measurements for different sensors indicated that, within an ozone concentration range of 10–1500 ppb, the sensitivity of coumarin 47–based sensors remained constant, while that of rodamin B–based sensors increased from 10 to 80 ppb, subsequently remaining constant. For sensors using coumarin 30 an increase in sensitivity was detected in the 10–60-ppb range. Thus, as a result of the studies performed, a sensor using a fine-grained borosilicate glass backing based on a coumarin 47–gallic acid compound was selected.

To investigate the chemiluminescence mechanism using a high-sensitivity spectrophotometer, the chemiluminescent spectrum of a rodamin B–based sensor was measured. This spectrum is presented in Fig. 3 along with that of photoluminescence.

The sensitivity of the sensor showed to be not influenced by humidity levels comparable to the stratospheric values.

3. Ozone analyzer design

The results obtained from the study of chemiluminescent sensors led to the planning of a fast ozone analyzer (FOZAN) to be used for the stratospheric measurements performed in the framework of the APE.

The ozone analyzer layout is presented in Fig. 4. Inside a chemical reactor, a chemiluminescent sensor is placed along a line where the air flow to be analyzed is provided by a pneumatic pump. The luminescence light produced by the sensor is modulated using an optical liquid-crystal chopper, controlled by a microprocessor unit, and registered by an electronic unit. The microprocessor unit not only regulates the optical modulator but also governs the operations of all the units included in the data processing.

A double-entry pneumatic valve connects the chemical reactor either with the built-in calibrating ozone generator or the air flow to be analyzed. At the input of the instrument, the airflow is thermostated with an heater and, on its way to the calibrating ozone generator, passes through an ozone destruction cell. Unit 10, reported in Fig. 4, comprises a pressure gauge and an automatic regulation system to maintain the ozone generator at constant ozone concentration and variable pressure.

According to the commands given by the controller, the instrument can operate in two modes: measurement or calibration.

In the measurement mode, the thermostated airflow is passed through the valve into the reactor, after which it undergoes the ozone-sensor chemiluminescence reaction. Based on the luminescence values and calibration results, the controller derives the absolute ozone concentration value, which is shown on a display panel.

In the calibration mode, the valve passes the airflow to the reactor through the ozone calibration cell and calibrating ozone generator. Following a set program, the controller registers dark and overall currents and computes the calibration coefficient. The calibration time is 70 s.

The two operational modes are switched from one to the other automatically.

4. Device specifications

The thermostatic unit that controls the input airflow maintains the temperature of the flow at a somewhat higher level than its expected maximum. A gear-type pneumatic pump (10 L m−1) provides the airflow, which is independent of pressure within the range 50–800 tor.

The comparator used in the thermostatic unit intercompares the generator’s sawtooth voltage with that of the thermal resistor installed in the unit. This intercomparison brings about a change in the relative pulse duration and the effective current passing through the spiral of the heater. To prevent ozone destruction, the spiral is covered with a Teflon sheath. The temperature stabilization error does not exceed 1%.

A similar system is applied to the temperature control in the quartz bulb of the UV lamp in the calibrating ozone generator. The power of the input thermostat is sufficient to stabilize the airflow at 70°C minimum temperature.

The calibrating ozone generator supplies an airflow with a strictly specified ozone concentration to the chemical reactor. The generator produces ozone due to the effect of UV radiation from a DRL-125-type mercury lamp on the airflow. The thermostatic control of the bulb at 60°C prevents the influence of varying air temperature and bulb heating with discharge current on the mercury vapor pressure and, hence, on UV radiation intensity.

The DRL-125-type mercury lamp is connected in series with a diode bridge that carries the load of an electric generator. The latter uses a high-voltage transistor as a regulation element and an operational amplifier providing feedback control. The discharge current of the mercury lamp is maintained at a given level by the electric generator, thus attaining the calibration level of ozone concentration.

To eliminate the effects of atmospheric pressure on the operation of the calibrating ozone generator, a pulse-width control of the duration of the mercury lamp discharge burning is used. The rate of switching DRL-125 is 15 Hz.

The instrument uses an ASCX15AN-type pressure gauge (SenSym, United States) whose signals are utilized by the comparator with the sawtooth voltage. As a result, depending on the level of the pressure-gauge signal, pulses with the corresponding length are formed and a stable calibrating ozone concentration is thus preserved.

To modulate chemiluminescent radiation, an optical liquid-crystal modulator is installed in front of a photomultiplier input window. It provides pulse modulation of the light flux at 1-Hz frequency and 0.5 relative pulse duration. The luminescence signals are registered by the FEU-100-type photomultiplier. Subsequently, the electric signals are passed to a dc amplifier, then to an analog-to-digital converter (ADC) to be transformed in a digital code, and finally, to the controller for further processing.

The ADC consists of an inverting amplifier based on a KR140UD20 with a gain coefficient of 2; an analog switch based on a KR590KN4; an integrator based on a K544 UD2B-type operational amplifier with high input resistance, providing high linearity of conversion; and a comparator based on a KR140UD20B.

The ADC converts analog signals to a 10-bit digital code at a required rate.

The microcontroller (KR580VM80A), which governs all operations of functional devices along with data processing, data transmission, and representation, has a clock period of 2 MHz.

The main technical characteristics of the prototype are summarized in Table 1.

5. Comparison results and discussion

A first calibration was performed utilizing a slow-speed response time Dasibi ozone analyzer model 1008 (Dasibi Environmental Corporation, United States), which had been previously compared with the Dasibi ozone calibration unit at the MicroPhysics Laboratory of FISBAT–CNR. A further comparison against an electrochemical cell (ECC) was performed to check the performances of the sensor over a wide ozone concentration range.

Differences less than 10% were recorded between Dasibi and FOZAN for ambient measurements over a long monitoring period (Fig. 5a). This calibration, performed in very similar environmental conditions was, obviously, only indicative of the reliability of the analyzer and represented the ground truth.

The result of the comparison with the ECC are reported in Fig. 5b. It is possible to note the very good agreement between the two sensors and the linearity of the signal produced by FOZAN.

Undoubtedly, the most important test of the instrument were the test flights carried out at the Pratica di Mare Air Force Airport, Italy, using the stratospheric aircraft M-55 Geophysica. Three test flights of about three hours each were performed. During each test flight the aircraft reached altitudes of about 20 000 m and temperatures lower than −70°C.

Some vertical ozone soundings, utilizing electro-chemical cells, were performed during the same period at Aquila University (situated at some 150 km from Pratica di Mare).

Figure 6 reports the vertical ozone profiles obtained by FOZAN and the balloon sounding on 13 November 1996. This test flight was chosen for the comparison because of its good correspondence with the balloon launch.

A detailed comparison reveals some remarkable patterns.

  • In the lower troposphere the two profiles have a completely inverse slope, the balloon instrument showing increasing values with height while those of FOZAN decrease.

  • Up to 10 000 m the slopes are very similar but the two profiles differ in magnitude, the balloon values being approximately double those of FOZAN.

  • From 12 000 m up to 20 000 m the balloon profile lies well within the data distribution obtained by FOZAN.

Differences in the troposphere results can be explained by the fact that the balloon and airplane were launched at two different locations, the first rural, the latter coastal, and thus encountered different air masses. Considering the planetary boundary layer only, ozone undergoes different transport and deposition phenomena at the two sites, which in turn leads to different concentration values.

FOZAN’s ozone concentration values obtained for the lower stratosphere are, on the average, in very good agreement with the values obtained by the balloon. For most of the ascent the two profiles also exhibit the same slope.

6. Conclusions

The new ozone analyzer FOZAN, specially designed for stratospheric measurements of ozone concentration, was found to be suitable for this kind of application when working in extreme environmental conditions during test flights performed on board Geophysica aircraft. The comparison of data recorded during one flight with the vertical profile recorded by a simultaneous balloon sounding showed that the instrument had a good reliability.

Acknowledgments

This work was funded by the PNRA, European Union Environmental Programs (DGXII), and ASI, and supported by ESF and APE Comitato di gestione. A special thanks to Dr. V. Rizi and Prof. G. Visconti for kindly supplying the balloon data. The authors are indebted to the PNRA Chairman Dr. Mario Zucchelli for his strong personal contribution in supporting this project.

REFERENCES

  • Bersis, D., and E. Vassilion, 1996: A chemiluminescent method for determining ozone. Analyst,91, 499–505.

    • Crossref
    • Export Citation
  • Güsten, H., G. Heinrich, R. W. H. Schmidt, and U. Schurath, 1992:A novel ozone sensor for direct eddy flux measurements. J. Atmos. Chem.,14, 73–84.

    • Crossref
    • Export Citation
  • Hodgeson, J. K., K. J. Krost, A. E. Keefe, and R. K. Stevens, 1970:Chemiluminescent measurements of atmospheric ozone. Analyt. Chem.,42, 1795–1802.

    • Crossref
    • Export Citation
  • Kramer, H. J., 1996: Observation of the Earth and its Environment. 3d ed. Springer-Verlag, 960 pp.

    • Crossref
    • Export Citation
  • Ray, J. D., D. H. Stedman, and G.J. Wendel, 1986: Feast chemiluminescent method for measurement of ambient ozone. Analyt. Chem.,58, 598–600.

    • Crossref
    • Export Citation
  • Sahand, S., W. Spenser, and U. Schurath, 1986: A battery powered light weight ozone analyser for use in the troposphere and stratosphere. Proc. Fourth European Symp. Phys-Chem. Behaviour Pollutants, Stresa, Italy, 33–44.

    • Crossref
    • Export Citation
  • Stefanutti, L., A. Adriani, R. Azzolini, S. Bormann, B. Carli, G. Fiocco, G. Giovanelli, T. Georgiadis, and V. Khatatov, 1995: Airborne Polar Experiment. Life Chem. Rep.,13, 57–67.

  • WMO, 1995: Scientific assessment of ozone depletion:1994. Rep. 37.

Fig. 1.
Fig. 1.

Activation curve for sensors with coumarin 47 (sensitivity vs time).

Citation: Journal of Atmospheric and Oceanic Technology 16, 10; 10.1175/1520-0426(1999)016<1345:ACAFSM>2.0.CO;2

Fig. 2.
Fig. 2.

(a) Kinetic characterization for sensors with coumarin 47 for different ozone concentration values. (b) Kinetic characterisation for sensors with rodamin B for different ozone concentration values.

Citation: Journal of Atmospheric and Oceanic Technology 16, 10; 10.1175/1520-0426(1999)016<1345:ACAFSM>2.0.CO;2

Fig. 3.
Fig. 3.

Spectrum of (1) photoluminescence and (2) chemiluminescence of rodamin B as function of wavelength.

Citation: Journal of Atmospheric and Oceanic Technology 16, 10; 10.1175/1520-0426(1999)016<1345:ACAFSM>2.0.CO;2

Fig. 4.
Fig. 4.

FOZAN: technical layout and specifications.

Citation: Journal of Atmospheric and Oceanic Technology 16, 10; 10.1175/1520-0426(1999)016<1345:ACAFSM>2.0.CO;2

Fig. 5.
Fig. 5.

(a) Comparison between FOZAN and Dasibi within the range 0–450 ppbv. (b) Comparison between FOZAN and ECC sensor within the range 0.5–3.5 ppmv.

Citation: Journal of Atmospheric and Oceanic Technology 16, 10; 10.1175/1520-0426(1999)016<1345:ACAFSM>2.0.CO;2

Fig. 6.
Fig. 6.

Comparison of FOZAN (ṡ) and balloon (continuous line) obtained during the test flight.

Citation: Journal of Atmospheric and Oceanic Technology 16, 10; 10.1175/1520-0426(1999)016<1345:ACAFSM>2.0.CO;2

Table 1.

Main technical characteristics of FOZAN.

Table 1.
Save
  • Bersis, D., and E. Vassilion, 1996: A chemiluminescent method for determining ozone. Analyst,91, 499–505.

    • Crossref
    • Export Citation
  • Güsten, H., G. Heinrich, R. W. H. Schmidt, and U. Schurath, 1992:A novel ozone sensor for direct eddy flux measurements. J. Atmos. Chem.,14, 73–84.

    • Crossref
    • Export Citation
  • Hodgeson, J. K., K. J. Krost, A. E. Keefe, and R. K. Stevens, 1970:Chemiluminescent measurements of atmospheric ozone. Analyt. Chem.,42, 1795–1802.

    • Crossref
    • Export Citation
  • Kramer, H. J., 1996: Observation of the Earth and its Environment. 3d ed. Springer-Verlag, 960 pp.

    • Crossref
    • Export Citation
  • Ray, J. D., D. H. Stedman, and G.J. Wendel, 1986: Feast chemiluminescent method for measurement of ambient ozone. Analyt. Chem.,58, 598–600.

    • Crossref
    • Export Citation
  • Sahand, S., W. Spenser, and U. Schurath, 1986: A battery powered light weight ozone analyser for use in the troposphere and stratosphere. Proc. Fourth European Symp. Phys-Chem. Behaviour Pollutants, Stresa, Italy, 33–44.

    • Crossref
    • Export Citation
  • Stefanutti, L., A. Adriani, R. Azzolini, S. Bormann, B. Carli, G. Fiocco, G. Giovanelli, T. Georgiadis, and V. Khatatov, 1995: Airborne Polar Experiment. Life Chem. Rep.,13, 57–67.

  • WMO, 1995: Scientific assessment of ozone depletion:1994. Rep. 37.

  • Fig. 1.

    Activation curve for sensors with coumarin 47 (sensitivity vs time).

  • Fig. 2.

    (a) Kinetic characterization for sensors with coumarin 47 for different ozone concentration values. (b) Kinetic characterisation for sensors with rodamin B for different ozone concentration values.

  • Fig. 3.

    Spectrum of (1) photoluminescence and (2) chemiluminescence of rodamin B as function of wavelength.

  • Fig. 4.

    FOZAN: technical layout and specifications.

  • Fig. 5.

    (a) Comparison between FOZAN and Dasibi within the range 0–450 ppbv. (b) Comparison between FOZAN and ECC sensor within the range 0.5–3.5 ppmv.

  • Fig. 6.

    Comparison of FOZAN (ṡ) and balloon (continuous line) obtained during the test flight.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 579 226 73
PDF Downloads 164 30 7