D1.
High-Sensitivity Measurements of Hydrocarbon Species Using Interband Cascade Lasers Operating Near 3.3 Microns
Mark G. Allen, David M. Sonnenfroh, and Seonkyung Lee
Physical Sciences Inc.
20 New England Business Center, Andover, MA 01810
allen@psicorp.com www.psicorp.com
Progress in Type-II, Interband Cascade Laser (ICL) technology has resulted in room-temperature semiconductor lasers operating at wavelengths near 3.3 microns. This wavelength region is technologically significant for high-sensitivity detection of a variety of hydrocarbon species such as methane, ethylene, and mixed vapors from gasoline.
This poster will present initial results using a single-mode Distributed FeedBack (DFB) ICL device to measure methane. The present DFB device is operated cw at liquid nitrogen temperatures, although room-temperature devices are expected later in 2003. The poster will describe the overall sensor configuration and discuss laser operating parameters such as threshold current, operating power, tuning range, and linewidth.
D2.
HIGH SENSITIVE 1.31 um DIODE LASER HYDROGEN FLUORIDE SENSOR WITH DETECTION LIMIT 1 ppb
A.G.Berezin, O.V.Ershov, A.I.Nadezhdinskii, Ya.Ya.Ponurovskii
Natural Sciences Center of A.M.Prokhorov General Physics Institute of Russian Academy of Sciences 119991 GSP-1 Vavilova st. 38, Moscow, Russia
The automated high sensitive laboratory prototype of hydrogen fluoride sensor was developed on a base of 1.31 um diode laser spectrometer operating in pulsed regime [1]. In the region 1.26 - 1.33 um hydrogen fluoride has rather strong absorption band (first overtone of the single fundamental band). There are plenty of commercially available diode lasers at wavelength 1.31 um that could reach several absorption lines of HF (P4 – 1.32125 um, P3 –1.31259 um , P2 – 1.30453 um), which could be used for detection. However, choice of diode lasers used for detection was rather difficult. Most convenient single-mode single-frequency distributed feedback (DFB) diode lasers should be carefully tested before applying to measurements, otherwise HF absorption line frequency may get into the dead intervals of their operation (intervals of mode hopping). Taking into account large distance between HF lines, which could be compared with diode laser tuning range, many of tested DFB lasers were useless for detection in spite of their single frequency (within one mode)operation. An employment of Fabry-Perout lasers instead of DFB lasers leads to another problem. These lasers have several modes with broader tuning range involving water lines. In this case it was difficult to distinguish between HF absorption line in main mode and additional absorption of water in satellite generation modes. Finally Fabry-Perout diode laser was chosen for detection of P2 line and water absorption was subtracted by data processing.
The device incorporated Chernin four-objective multipass optical cell with optical path length 39 m at base length of 25 cm. Electronics controlling the device along with single-mode-fiber-coupled diode laser and reference cell were mounted in separate box, connected to multipass cell via optical fiber and to computer via electric cable. The device was run by computer program written in LabVIEW-6. The device was rechargeable battery powered; the energy supply was enough for 8 hrs of continuous operation. The main destination of the sensor was the monitoring of HF content in the ambient air. Std limit for 0.4 s averaging was found to be less than 1 ppb, which is quite acceptable for a number of applications.
[1] Nadezhdinskii A., Berezin A., Chernin S., Ershov O., Kutnyak V. Spectrochimica Acta Part A, 55, 2083 (1999).
D3.
RAPID PASSAGE AND POWER SATURATION EFFECTS IN PULSED QUANTUM CASCADE LASER SPECTROMETERS
M.T. McCulloch, G. Duxbury and N.Langford
Department of Physics, John Anderson Building, University of Strathclyde,
107 Rottenrow, Glasgow G4 0NG, Scotland,UK
Recently their has been considerable interest in adiabatic passage effects in atomic and molecular gases subject to short laser pulses. In the infrared region the relation times of low pressure molecular gases are on the microsecond time scale. In our experiment the time of passage of the chirped QCL pulse through a Doppler broadened line is sub-ns, very much faster than in earlier experiments, greatly enhancing the chance of seeing rapid passage effects. Since the intensity of the pulse is about, 104 W m-2, the combination of high intensity and a short interaction time, which is much faster than the relaxation processes, leads to the observation of strong adiabatic rapid passage signals and power dependent bleaching. We have observed these effects in several gases, in particular ethylene, methyl chloride and ammonia.
Examples are given of the effects of the short interaction time and power bleaching on the observed spectral profiles of lines, even when pressure broadened with up to 100 Torr of nitrogen. Although only three molecular examples are considered in detail, we will show that these effects are present when any short, intense QCL pulse is used to probe the spectrum of specific molecules such as ethylene, irrespective of the mode of operation of the spectrometer.
D4.
Process Gas Analysis by Infrared Spectroscopy in the Semiconductor Industry
L. Emmenegger*, J. Mohn
Swiss Federal Materials Testing and Research Laboratories, CH-8600 Dübendorf
*corresponding author: Dr. Lukas Emmenegger phone: +41 1 823 46 99;
fax: +41 1 821 62 44 email: lukas.emmenegger@empa.ch
A wide range of speciality gases are used in the production of microelectronic devices, such as CMOS, LEDs and TFTs. These gases include mainly (per)fluorocompounds (e.g. CF4, WF6, NF3), silanes (SiH4, [(C2H5O)4Si]), and dopants (AsH3, PH3). The use and abatement of these gases must be optimised because of their high costs, agressive properties and strong global warming potential. The present study includes chemical vapour deposition (CVD) of tungsten and silicon dioxide, with special focus on the gaseous reactions and mass balances during cleaning steps.
Extractive on-line FTIR was used for qualitative and quantitative analysis of the main gaseous products. Materials and analytical parameters were optimized to achieve rapid instrumental response at high spectral resolution. Quantification was done by classical least-square algorithms. The measurement of adequate refererence spectra was one main challenge because of the reactive and uncommon gases studied.
The results indicate large optimization potential for some process steps. The quantitative results and spectral information will be valuable for the development of enhanced process control and adequate analytical methods. FTIR has proven to be a powerful tool for overall studies. The results obtained will promote future, more specific developments of analytical systems - including TDLS - for process gas analysis in the semiconductor industry.
D5.
MEASUREMENTS OF RELATIVE INTENSITY NOISE OF QUANTUM CASCADE LASERS
T. Gensty, W. Elsäßer
Institute of Applied Physics, Darmstadt University of Technology, Schlossgartenstrasse 7,
D-64289 Darmstadt, Germany
We present measurements of the relative intensity noise (RIN) of quantum cascade (QC) lasers. The RIN is analyzed for free-running QC lasers and for QC lasers under optical feedback. Since their first realization in 1994 [1], QC lasers have made tremendous progress. Recently, continuous wave (cw) operation has been reported at room-temperature [2]. Because of the high operation temperatures and the high output power in the mid-infrared spectral region, QC lasers have already become suitable light sources for trace gas sensing and for optical free-space communication. So far, only little attention has been paid to the intensity noise properties of these novel devices.
Here, we present investigations of the RIN of Fabry-Perot QCLs [3] and DFB QC lasers. Figure 1 depicts the RIN* (measured under quasi-cw operation) of three different QC lasers.
These results are analyzed with a single mode rate equation noise approach taking into account the different device parameters, possible multi-mode contributions and the particular difference between QC lasers and edge emitting lasers and vertical-cavity surface-emitting lasers [4]. Finally, we introduce controlled optical feedback and compare the RIN results with those of the free-running case. Possible consequences for measurement system applications will be discussed.
[1] J. Faist et al., Science 264, 553 (1994).
[2] M. Beck et al., Science 295, 301 (2002).
[3] Q. K. Yang et al., Appl. Phys. Lett. 80, 2048 (2002).
[4] D. M. Kuchta et al., Appl. Phys. Lett. 62, 1194 (1993).
Share with your friends: |