328 Propellants, Explosives, Pyrotechnics 23, 328Ä…332 (1998)
Radiative Ignition of Pyrotechnics: Effect of Wavelength on Ignition
Threshold
Leo de Yong and F. Lui
Defence Science and Technology Organisation, Weapons Systems Division, Aeronautical and Maritime Research
Laboratory, Melbourne VIC 3001 (Australia)
Strahlungszundung von Pyrotechnica: Die Wirkung der Wellen- Amorcage par rayonnement de produits pyrotechniques: effet de
È Ë
È È Ë
lange auf die Zundschwelle la longueur d'onde sur le seuil d'amorcage
Á Â Á
Unter Verwendung einer Btreitband-Xenonlampe und einer leis- En utilisant une lampe a xenon a large bande et une diode laser
È È Â Â Ë Â
tungsfahigen Laserdiode wurde die Zundung verschiedener pyro- performante, on a etudie l'amorcage de differentes compositions
È
technischer Satze unter Einwirkung von Strahlung im Ultraviolett, im pyrotechniques sous l'effet d'un rayonnement dans l'ultraviolet, dans
È Â Â Â
Sichtbaren und im Infrarot untersucht. Bei Strahlenstarken von le visible et dans l'infrarouge. Avec des intensites energetiques de
È È Â Â Â
4,9 W=cm2 wurde durch UV-Strahlung keiner der Satze gezundet, 4,9 W=cm2, aucune des compositions n'a ete amorcee. En revanche,
 Á  Á
dagegen reagierten SR112, B=Fe2O3 und Schieûpulver auf sichtbare SR112, B=Fe2O3 et la poudre a canon ont reagi a un rayonnement
È Â Â Â
Strahlung bei einer niedrigen Strahlenstarke von 8,1 W=cm2 Die visible de faible intensite energetique (8,1 W=cm2). La plupart des
È È Â Â Â
meisten Satze wurden durch IR-Strahlung gezundet, jedoch konnten compositions ont ete amorcees par un rayonnement IR, mais
È Ã Â Ã
Mg=NaNO3 und SR112 selbst bei Strahlenstarken bis 300 W=cm2 Mg=NaNO3 et SR112 n'ont pu etre amorcees meme avec des inten-
nicht gezundet werden. Es wird angenommen, daû der Unterschied im sites energetiques atteignant 300 W=cm2. On suppose que la differ-
È Â Â Â Â
È Â
Ansprechverhalten der Satze nicht in direkter Beziehung zu ihrer ence de reaction des compositions n'est pas en relation directe avec
È Â Â
Strahlenabsorption steht, aber abhangig ist von vielen physikalischen, leur capacite d'absorption de rayonnement, mais depend de nom-
 Â
chemischen und optischen Eigenschaften der pyrotechnischen Pulver. breuses proprietes physiques, chimiques et optiques des poudres
È È È Â Â Â Â
Zunahme der Strahlenstarke reduziert die Zundzeit fur die meisten pyrotechniques. Une augmentation de l'intensite energetique reduit le
È Ë
Satze. temps d'amorcage de la plupart des compositions.
Summary cess(1Ä…9). Highly reliable RF=ESD=EMP immune igniters
have been built using miniature lasers, high power laser
Using a broadband Xenon lamp and a high power laser diode, the
diodes or optically pumped laser rods. The attainment of
ignition of several pyrotechnic compositions was evaluated using
high power output has also led to the use of lower sensi-
ultraviolet (UV), visible and infrared (IR) radiation. At irradiance
tivity igniter compositions further increasing the safety of
levels of 4.9 W=cm2 none of the compositions were ignited using UV
radiation but SR112, B=Fe2O3 and gunpowder showed reactions to ignition systems. The concept of laser diodes coupled with
visible radiation at irradiance levels as low as 8.1 W=cm2. Most of the
®ber optics has also led to novel and versatile igniter
compositions were ignited with IR radiation but Mg=NaNO3 and SR
design(10Ä…15). However, although lightweight, high power
112 could not be ignited at irradiance levels of up to 300 W=cm2. It is
laser diodes are still relatively costly.
proposed that the difference in the response of the compositions was
not directly related to the radiation absorption characteristics of the Whilst many studies into alternative ignition sources for
compositions; but was dependant on many physical, chemical and
pyrotechnics have examined the use of IR radiation, none
optical characteristics of the pyrotechnic powders. Increasing the
have examined the effects of visible or ultraviolet (UV)
irradiance reduces the time to ignition for most compositions.
radiation. Many UV and visible radiation sources are
available at relatively low cost, have high power output and
they may offer cheap alternatives to IR sources for pyro-
technic ignition.
1. Introduction
This preliminary study explores the use of broadband
UV, visible and IR radiation as alternative ignition sources
Ignition of relatively simple pyrotechnic systems or
for some common pyrotechnic compositions. It also uses a
devices is usually initiated by a percussion mechanism
high power laser diode as a narrow band comparative
(including stab) or by a friction process. However, more
radiation source.
complex systems (e.g., rocket motors etc.) use electrically
initiated pyrotechnic igniters which are usually low voltage
devices. But the use of low voltage igniters in the modern
military environment has led to the problem of inadvertent
2. Experimental
initiation of the igniter due to stray electrostatic discharge or
spurious RF signals. Designing to a 1 amp=1 watt criteria or
shielding the igniter may reduce the problem but it is 2.1 Xenon Lamp
usually at the expense of increased cost and=or weight.
Over the last twenty years, the concept of using infrared A Cermax Xenon LX300 UV light source with a spectral
(IR) radiation as an ignition source for energetic materials output as shown in Fig. 1 was used. The broadband power
has been extensively studied with a large degree of suc- was measured as 30 W whilst that in the individual UV,
# WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998 0721-3115/98/0306Ä…0328 $17.50‡:50=0
Propellants, Explosives, Pyrotechnics 23, 328Ä…332 (1998) Effect of Wavelength on Ignition Threshold 329
Figure 3. Schematic diagram of the laser diode experimental system.
diode with a maximum power output of 250 mW at 810 nm
Figure 1. Spectral output of the Cermax Xenon UV lamp.
was used. The diode as supplied was pigtailed and, after
splicing to a 100 mm step index optical ®ber, the maximum
visible and IR portions of the lamp output were 4.7 W,
output was reduced to 210 mW. Several arrangements of the
7.8 W and 17.5 W, respectively.
®ber and the sample were tested: the ®ber touching the front
The experimental arrangement used to measure the
face of the pressed composition, the ®ber touching a 0.5 mm
response of the pyrotechnic compositions is shown in Fig. 2.
BK7 glass window onto which the composition had been
The output of the lamp was focussed onto the surface of the
pressed and the ®ber positioned a small distance away from
sample and thermal paper was used to measure the spot
the front face of the pressed composition. The most suc-
size. The spot diameter was approximately 11 mm and its
cessful arrangement was to position the ®ber 0.5 mm from
pro®le was approximately Gaussian. Three ®lters were used
the front face of the sample. This gave a spot diameter of
to separate the light source output into either the UV (UG-
approximately 300 mm at the sample. The sample was
11, 90% transmission at peak wavelength of 340 nm),
irradiated with a 300 ms pulse of radiation and the response
visible (BG-40, 98.5% transmission at peak wavelength of
of the sample was recorded with a second optical ®ber
500 nm) or the IR (RG665, 99% transmission at peak
positioned next to the input ®ber. The difference between
wavelength of 710 nm) portions of the spectrum. The dis-
the start of the input laser pulse and the response of the
tance between the sample and the lamp was varied slightly
sample was recorded as the time-to-ignition.
for each ®lter arrangement to ensure that the maximum
irradiance (minimum spot size) occurred at the sample
surface. Initially, the power output of the lamp was set to
2.3 Pyrotechnic Compositions
maximum and the sample irradiated for 10 s using a
mechanical shutter to control the exposure time. If the
Table 1 lists the pyrotechnic compositions studied. For
composition ignited, the exposure time was decreased in
the Xenon lamp tests, 30 mg of the composition was pressed
regular steps until ignition did not occur. The time-to-
into a metal cup at 16.5 MPa. For the laser diode tests,
ignition was recorded as the minimum exposure time to
1000 mg of composition was pressed into a 5.8 mm dia-
achieve sample ignition. Due to the variable response of
meter perspex holder at 6.7 MPa.
some of the samples, the recorded time-to-ignition was
sometimes broad.
2.4 Composition Spectra
2.2 Laser Diode
UV and visible absorption spectra of each composition
were measured with a Varian Carey 3 Spectrophotometer.
The experimental setup shown in Fig. 3 was used to
measure the response of the compositions to narrow band IR
Table 1. Details of Pyrotechnic Compositions Studied
radiation. An SDL-2430-H2 (Spectra Diode Labs.) laser
Composition Designation Proportions (% w=w)
Mg=NaNO3 Ð 50 : 50
Mg=NaNO3=ZnO Ð 47.5 : 47.5 : 5
Mg=NaNO3=ZnO Ð 45 : 45 : 10
B=Fe2O3 MRL(X)201 25 : 75
B=Fe2O3=ZnO Ð 23.5 : 71.5 : 5
B=Fe2O3=ZnO Ð 22.5 : 67.5 : 10
Gunpowder G20 Ð
B=KNO3 SR 44 30 : 70
B=KNO3 SR 43 50 : 50
TNC(a)=KNO3 SR 112 40 : 60
Si=KNO3=SMP(b) SR 252 40 : 40 : 20
(a) Tetranitrocarbazole
Figure 2. Schematic diagram of the Xenon lamp experimental
(b) Sulphurless Mealed Powder
system.
330 Leo de Yong and F. Lui Propellants, Explosives, Pyrotechnics 23, 328Ä…332 (1998)
Table 2. Times-to-Ignition for Pyrotechnic Compositions Exposed to UV, Visible and IR Radiation
Composition Time-To-Ignition (ms)
UV Radiation Visible Radiation IR Radiation Total Radiation Laser Diode(c)
Mg=NaNO3 N(a) N N N N
Mg=NaNO3 ‡ 5% ZnO N N N N N
Mg=NaNO3 ‡ 10% ZnO N N N N N
B=Fe2O3 N 10000 400Ä…500 120Ä…130 7.5
B=Fe2O3 ‡ 5% ZnO N N 200Ä…300 200 -
B=Fe2O3 ‡ 10% ZnO N N 1000 120 -
SR 43 N N 750Ä…1000 300Ä…400 -
SR 44 N N 800Ä…900 150Ä…170 60.0
SR 112 N P N 600Ä…700 N
SR 252 N N 900Ä…1500 170Ä…180 -
G 20 N P(b) 250Ä…300 90Ä…100 -
(a) N designates no ignition
(b) P designates partial ignition
(c) The time-to-ignition recorded here is that measured at an irradiance of 200 W=cm2 for comparison purposes
IR absorption spectra of each composition were measured Irradiation with the IR portion of the output (irradiance
with a Bruker ISF88 FTIR Spectrophotometer using a KBr 18.5 W=cm2) resulted in successful ignition of many of the
disc. compositions with the exception of the Mg=NaNO3 and SR
112 formulations. IR spectra of the compositions showed
strong absorption for all the compositions with the excep-
tion of B=Fe2O3, Mg=NaNO3 and SR 112. Ignition times
3. Results ranged from 200 ms (B=Fe2O3=ZnO) to 1000 ms (SR 43).
Attempts to ignite the compositions using the combined
The results for the ignition of the pyrotechnic samples UV, visible and IR radiation from the lamp (irradiance
using the Xenon lamp are detailed in Table 2. 31.5 W=cm2) resulted in ignition of all compositions except
None of the compositions could be ignited with the UV those based on Mg=NaNO3. Times-to-ignition varied from
portion of the lamp output (irradiance 4.9 W=cm2) even 90 ms (G20) to 700 ms (SR 112).
after exposure for up to 10 s. All the compositions showed Ignition with the laser diode was evaluated with a limited
good absorption of UV radiation except SR 252 (Fig. 4) and number of compositions. As observed with the Xenon lamp,
although the addition of 5% and 10% ZnO slightly the Mg=NaNO3 and SR 112 compositions could not be
increased the UV absorption for the Mg=NaNO3 and ignited with the laser diode even at the maximum irradiance
B=Fe2O3 compositions, it did not result in ignition of the of 300 W=cm2. However, SR 44 and B=Fe2O3 compositions
samples. were ignited with threshold laser irradiances of approxi-
Attempts to ignite the compositions with visible radiation mately 39.0 W=cm2 and 50.0 W=cm2, respectively. For
(8.1 W=cm2) were more successful with complete reaction these tests, Fig. 5 shows that as the irradiance increases, the
of B=Fe2O3 and partial ignition of SR 112 and G 20 being time-to-ignition decreases to a limiting minimum time.
observed after exposure for 10 s. Most of the compositions Conversely, as the irradiance decreases, the time-to-ignition
strongly absorbed the visible radiation; Mg=NaNO3 showed
the weakest overall absorption and SR 112 exhibited strong
absorption in the waveband 400Ä…500 nm but decreasing
weak absorption from 500Ä…700 nm (Fig. 4).
Figure 4. UV and visible absorption spectra for several pyrotechnic Figure 5. Time-to-ignition measured as a function of laser diode
compositions. irradiance for B=Fe2O3 composition.
Propellants, Explosives, Pyrotechnics 23, 328Ä…332 (1998) Effect of Wavelength on Ignition Threshold 331
increases to the limiting irradiance where ignition will not role in the ignition process for this composition. The results
be achieved. obtained with the laser diode tend to con®rm this observa-
Ignition with the laser diode was strongly dependant on tion as ignition of SR 112 could not be achieved even at IR
the physical arrangement of the sample and the optical ®ber. irradiances up to 300 W=cm2 presumably due to the poor
Using the B=Fe2O3 composition and with the ®ber or the absorption of the IR radiation. However, Table 2 suggests
glass window in contact with the composition, the values of that for most of the other compositions it is the IR portion of
the time-to-ignition were very irregular, often varying by the radiation that is the predominant factor in the ignition
several hundred milliseconds for identical samples. Exam- process. This may simply be due to the greater available
ination of the ®ber and the window after many tests showed irradiance of the IR radiation compared to the UV and the
that part of the pyrotechnic composition had ignited and visible portions and=or to the relative difference in the
deposited reaction products either on the tip of the ®ber or optical properties of the compositions in the various regions
on the window. These deposits blocked all or part of the of the spectrum and the ef®ciency of heat generation in the
output of the ®ber and reduced the input power to the rest of molecules.
the sample. The effect was random and was overcome by The results also show that for the Xenon lamp tests,
placing the ®ber a ®xed distance from the front face of the irrespective of the spectral nature of the radiation, increas-
pressed composition with no window present. ing the irradiance from 8.1 W=cm2 to 18.5 W=cm2 to
31.5 W=cm2 resulted in a consistent decrease in the time-to-
ignition for all the compositions. Although similar beha-
4. Discussion viour was recorded with the laser diode, Fig. 5 shows that
there are limits that de®ne the minimum or threshold time-
The ignition of these pyrotechnic compositions appears to-ignition and irradiance for each composition.
to be dif®cult to achieve with UV radiation at the irradiance The Mg=NaNO3 compositions could not be ignited even
levels up to 5 W=cm2. It is expected that the UV energy will at irradiance up to 300 W=cm2 suggesting that ignition of
become delocalised as thermal energy within a system this composition will only be achieved at very high irra-
undergoing non-radiative energy relaxation. As all the diances.
compositions showed good absorption of the UV radiation, The problems with the experimental setup of the laser
it is likely that either the energy became too delocalised diode, optical ®ber, glass window and the sample were
within interatomic bonds of the oxidant to initiate decom- principally due to the nature of the compositions examined.
position and subsequent reaction with the fuel, or the irra- All the initial work was conducted with B=Fe2O3 which,
diance level was simply too low. when ignited, produces essentially gasless (liquid) reaction
Visible radiation successfully ignited three compositions products. As the products cool, the liquids condense and
but the times-to-ignition were long and sustained combus- adhere to the ®ber or the glass window. As noted earlier,
tion was only observed for one composition. There were this blocks part of the incoming radiation in an entirely
differences in the magnitude and the spectral character of random manner. It is expected that this type of problem
the visible radiation absorbed between those compositions would not occur with a more traditional ``gassy'' igniter
that ignited and those that did not (Fig. 4), but any simple composition.
relationship between the radiation absorption and ignition is
dif®cult to propose on the limited data presented here.
Exposure to IR radiation from the Xenon lamp resulted in
ignition of all the compositions except those based on
Mg=NaNO3 and SR 112. The failure of these two compo-
5. Conclusions
sitions to ignite may be due to their poor IR absorption of
the radiation. However, B=Fe2O3 also showed similar poor
IR radiation absorption yet was successfully ignited. (1) It has been successfully demonstrated that some pyro-
The ignition of a pyrotechnic composition by UV, visible technic compositions may be ignited using visible and
or IR radiation is a complex process. It will be dependant on infrared (both broadband and narrow band) radiation.
a combination of the physical=chemical=thermal and optical But, at an irradiance of 4.9 W=cm2, none of the com-
properties of the composition such as the temperature of positions could be ignited with UV radiation.
ignition, thermal diffusivity, heat of reaction, the radiation (2) The Mg=NaNO3 and SR 112 compositions could not be
absorption coef®cient, and the conversion of the absorbed ignited with IR radiation at power densities up to
radiation into heat by the molecules due to electronic 300 W=cm2 but SR 112 was partially ignited with
transitions or vibrational relaxation. Whilst it is not possible visible radiation at an irradiance as low as 7.7 W=cm2.
to compare the properties of each composition examined (3) For most of the pyrotechnic compositions the IR portion
here (as they were not measured), the results indicate that of the radiation was the predominant factor in suc-
the spectral properties of the radiation used are important cessful ignition being achieved. This may have simply
for ignition. been due to the greater irradiance of the IR radiation
For example, the results for SR 112 suggest that it is the compared to the UV or greater delocalisation of the
visible portion of the radiation that plays the more important thermal energy with the UV radiation.
332 Leo de Yong and F. Lui Propellants, Explosives, Pyrotechnics 23, 328Ä…332 (1998)
(7) L. de Yong and F. Valenta, ``A Study of the Radiant Ignition of a
(4) The spectral content of the radiation is important in
Range of Pyrotechnic Materials Using a CO2 Laser'', Report
ignition; SR 112 ignited at low visible irradiance levels
MRL-TR-90-20 (1990), Materials Research Laboratory, Mel-
but not at high IR irradiance levels.
bourne, Australia.
(5) The difference in the response of the compositions to (8) D. Ewick, L. Dosser, S. McComb, and L. Brodsky, ``Feasibility
of a Laser Ignited Pyrotechnic Device'', 13th International Pyro-
either UV, visible or IR radiation was not directly
technics Seminar, Grand Junction, July 1988.
correlated with the measured absorption spectra. For all
(9) D. Kramer, E. Spangler, and T. Beckman, ``Laser Ignited
the compositions, the absorption=reŻection of the
Explosive and Pyrotechnic Components'', American Ceramic
Society Bulletin 22, 78Ä…84 (1993).
radiation did not appear to be the dominant factor in the
(10) J. Petrick, ``Laser Ignition System Design for Pyrotechnic
ignition process.
Applications'', Proceedings of the American Preparedness
(6) For SR44 and B=Fe2O3 compositions, a threshold
Association Annual MeetingÄ…Pyrotechnics Section, Shreveport,
irradiance and threshold time to ignition were observed
October 1988.
(11) S. Kunz and F. Silas, ``Diode Laser Ignition of High Explosives
when exposed to narrowband IR radiation.
and Pyrotechnics'', 13th International Pyrotechnics Seminar,
Grand Junction, July 1988.
(12) R. Jungst, F. Silas, R. Watkins, and L. Kovucic, ``Development of
Diode Laser Ignited Pyrotechnic and Explosive Components'',
6. References
15th International Pyrotechnics Seminar, Boulder, July 1990.
(13) M. Landry, ``Laser Used as Optical Sources for Initiating
(1) V. Menichelli and L. Yang, ``Sensitivity of Explosives to Laser
È È
Explosives'', 16th International Pyrotechnic Seminar, Jonkoping,
Energy'', Report NASA JPL TR 32-1474 (1970), National
June 1991.
Aeronautics and Space Administration, Jet Propulsion Labora- (14) M. Landry and J. Cobbett, ``Laser Ordnance System (LOIS) for
tory, CA, USA.
Aircraft Egress Systems'', 16th International Pyrotechnics
(2) H. Oestmark, ``Laser as a Tool in Sensitivity Testing of Explo- Seminar, Jonkoping, June 1991.
È È
sives'', 8th International Symposium on Detonation, Albu- (15) B. Purdy, M. Fratta, and C. Boucher, ``Laser Ordnance System
querque, July 1985.
for NRL's ARTS Program'', 29th AIAA=SAE=ASME=ASEE
(3) J. Holy, ``Laser Initiation of TiHx=KClO4'', 11th International
Joint Propulsion Conference, Monterey, June 1993.
Pyrotechnics Seminar, Vail, July 1986.
(4) J. Holy and T. Girman, ``The Effect of Pressure on Laser
Acknowledgements
Initiation of TiHx=KClO4 and Other Pyrotechnics'', 13th Inter-
The authors would like to thank Lance Redman, for help with some
national Pyrotechnics Seminar, Grand Junction, July 1988.
of the experiments, Brian Jones, Kevin Lynch and Peter Collins for
(5) F. Al-Ramadhan, I. Haq, and M. Chaudri, ``Low Energy Laser
help with the instrumentation and David Hatt for advice and assistance
Ignition of Magnesium-TeŻon-Viton Compositions. J. Phys. D:
with the experimental hardware.
Appl. Phys. 26, 880Ä…887 (1993).
(6) B. Fetherolf, P. Liiva, W. Hsieh, and K. Kuo, ``CO2 Laser
Ignition Behaviour of Several Pyrotechnic Mixtures'', 16th
È È
International Pyrotechnics Seminar, Jonkoping, June 1991. (Received April 3, 1997; Ms 22=97)