Home   Background   Products  Applications  News  Contacts

LED-PD pairs in gas sensing

        Optically coupled mid infra-red light-emitting diodes and photodiodes are the most advanced elements for the creation of NDIR gas sensors. Small size and high brightness enables achieving long optical paths necessary for detection of gases in the low concentration range. Fig.1 presents dependence of the photosignal on distance between LED and PD equipped with immersion lenses of the Sr/Su type (D=3.5 mm). The LED current was fixed at 10 or 600 mA (pulse mode). On increasing the distance between LED and PD the signal decreases as ~(L)^-2.015 as shown in Fig.1 by a dotted line. For the convenience of comprehension the photodiode current was normalized so that both data fit one function.

Fig.1 

            As seen from Fig.1 at paths of about 1 m the signal –to-noise ratio (SNR) is still above unity that allows gas concentration measurements. However, long paths are achieved at a cost of sufficient reduction of the SNR that is followed by an accuracy decrease. That’s why the above optical scheme is used mainly for short paths, say 1-5 cm ( see right top of Fig.1 and Fig.9 in M A Remennyi et al ,Sensors & Actuators B: Chemical, Volume 91, Issues 1-3 , 1 June 2003, Pages 256-261 ).

            Utilization of diodes equipped with Si or chalcogenide glass microlenses (D=1 mm) allows better coupling of the diodes with total coupling efficiency as high as 70% [[i]] ( see Fig.2). Small LED dimensions allow efficient collimation of the radiation beam as well – the technology used in the open path instrument with optical path as long as 100 m [ [ii]  ] ( see Fig.3).

Fig.2                                                                                     Fig.3

            The analytical description of a NDIR gas sensor has been suggested (see Fig.2, 4 and ref. [[iii]]) with the aim to determine sensor instrument and transfer functions for estimating its main parameters as a measurement device. Such approach permitted us to produce the analytical model of the NDIR gas sensor that serves as a base for expert evaluations when comparing sensors with different hardware components and when solving applied problems of designing of different-purpose optical gas analyzers. The study presents the model of a CО2-sensor based on immersion diode optopairs. The calculated values of the threshold sensitivity at the level equal to tens of ppm, sensitivity not less than 0.1 vol.% and measurement accuracy at the level 1-2% within the range 0-10 v/v % СО₂ for the operational speed up to 10 readings per second and gas cell length L = 4 cm exceed the parameters of present-day produced CО2-sensors. The known pocket-size high-speed sensors that are used mainly in capnography have the sensitivity in the low-concentration range (up to 5 v/v % СО2) not less than 0.25 v/v % and relative accuracy of reading not less than 5% when measuring CO2 concentrations higher than 5 v/v % [[iv], [v], [vi]]. The experimental results of the study of the carbon dioxide sensor model confirm the validity of the model and fair prospects for using the sensors based on immersion diode optopairs in portable gas analyzers.

.