=Paper=
{{Paper
|id=None
|storemode=property
|title=Influence of temporal laser pulse length and shape on the time resolved laser induced incandescence signal
|pdfUrl=https://ceur-ws.org/Vol-865/Ditaranto.pdf
|volume=Vol-865
}}
==Influence of temporal laser pulse length and shape on the time resolved laser induced incandescence signal==
<pdf width="1500px">https://ceur-ws.org/Vol-865/Ditaranto.pdf</pdf>
<pre>
    Influence of temporal laser pulse length and shape on the
        time resolved Laser Induced Incandescence signal

                  M. Ditaranto, N.E. Haugen, C. Meraner, I. Saanum
    SINTEF Energy research, 7465 Trondheim, Norway, e-mail: mario.ditaranto@sintef.no

         Q-switched Nd:YAG lasers are typical choice for LII, with excitation pulse
length of 7-10 ns. The question of longer pulse duration was raised in Schulz, Kock
et al. (2006) and there is interest in using CW laser for LII (Black, 2010). The preset
study is an experimental investigation on the influence of pulse lengths, in the range
50 - 1000 ns, and temporal shapes. The measurements were made 25 mm above a
laminar non-premixed ethylene flame, using a Nd:YAG laser with temporal shaping
capabilities (Agilite, Continuum). The time resolved LII measurements were made by
re-constructing averaged, sequentially delayed, gated (10 or 20 ns) ICCD images in
radial profile or spectral modes. Except in spectral mode, the LII signal is recorded
with a 10 ns bandpass filter centered at 488 nm and a 532 nm centered notch filter.
By using the shaping capabilities of the laser the effect of pulse length has been
varied by keeping constant either the pulse fluence or pulse energy.
         The main feature that has been observed is shown in figure 1. At the start of
the pulse, the LII signal builds up equally for all cases, as a result of particles
absorbing energy and heating up. The LII signal for the 50 ns pulse is as expected,
decaying at a rate dependent on the primary particle diameter, but when the pulse
length is increased, one observes a shouldering of the signal after 50 ns (green
curve). This observation is unexpected as for constant pulse length, increasing the
fluence, is known to increase the decay rate (i.e. to narrow the LII signal), because of
the increase in vaporization rate (Michelsen, Witze et al.). It therefore indicates that
the effect of laser fluence expressed in J/cm2 has a time scale dependency on the LII
processes. A further increase in pulse length to 200 ns (red curve) shows not only a
further delay in the decay, but also a rebound of the LII signal. The cause of this
phenomenon is unclear, as none of the processes known to be involved in the LII
theory predicts such an effect. One shortcoming of our set up is that the beam is
formed into a sheet, however the spatial resolution is clearly defined by the collection
optics and this phenomenon would happen at all pulse lengths if it was due to the
interferences from out of focus signals. Figure 2 shows that the signal rebound
phenomenon as pulse length increases is also observed at constant fluence, starting
for pulses longer than ca. 75 ns. The behaviour was also detected with both a top hat
and a quasi-gaussian pulse temporal profile. The spectrally and time resolved LII
traces shown in figure 3 seem to exclude the effect of interferences from molecular
excitation processes such as late fluorescence, and confirm the efficiency of the
filtering scheme chosen in the measurements shown in fig. 1 and 2.




Fig. 1. Red: 200ns/100mJ; Green:   Fig. 2. E: 100mJ. Black: 600ns;   Fig. 3. Time resolved spectra taken at
100ns/45mJ; Blue: 50ns/30mJ.       Red: 200ns; Green: 100ns; Blue:   LII time shown in graph below.
                                   50ns.




                5th international workshop on Laser-Induced Incandescence
                 May 9-11, 2012, Palais des Congrès, Le Touquet, France

</pre>