In order to test the effectiveness of the gas - generating agents of airbag inflators to be used in
motor vehicles, it is important to measure the gas generation behavior within a closed vessel
that closely simulates the airbag. The 60 L - tank test is one of the most widely used methods.
However, 60 L - tank test apparatus are not very common in institutes other than automobile
companies and explosives manufacturers. In addition, there is a need for a relatively large
amount (20 g) of sample for each test.
In this study, we devised a new method of screening test of gas - generating agents, utilizing a
closed chimney - type strand burner. Compared to the conventional methods, the new method
have the merits in that : 1) there is no need to additionally purchase 60 L - tank test equipments ;
2) it requires less amount of samples compared to the 60L - tank test (total sample weight : 60 L -
tank test 20 g ; new method 4 g) ; 3) it is closer to the 60 L - tank test in volume than the smallscale
deflagration test (volume : smamm scale deflagration test 52 mL ; mew method 4 L) ; and 4) it
is easy to observe the burning behavior of the pellet.
To test the validity of the method, we compared the results of the rate of pressure rise inside
the vessel with those of the 60 L - tank test for compositions comsisting of tetrazole / oxidizer mixtures.
In the screening test, the composition to be tested was prepared by sieving and drying powder
samples of tetrazole and oxidizer, then mixing at a designated fuel / oxidizer - weight ratio by the
use of a rotary mixer. The composition was then pressed into a 4 g pellet. Unlike the linear
burning rate measurement, no restricting agent was applied at the bottom and at the sides of
the pellet. After mounting the pellet inside the vessel, N2 gas was introduced. The vessel was
closed, after checking that the operational pressure reached the designated level. The deflagration
of the pellet was initiated by the use of a nichrome wire. The pressure change inside the strand
burner after ignition was measured employlng a strain - gauge pressure sensor. The deflaglation
behavior of the pellet was also monitored by a video camera.
Comparing the results between the rate of pressure rise inside the strand burner and the
pressure rise after 50 ms, together with the rate of pressure rise inside the tank of the 60 L - tank
for the tested compositions, there were linear relationships for gas - generating agents whose
fuel are composed of different tetrazoles, indicating good correlation. There was also a linear
relationship for the result of the rate of pressure rise between the strand burner and the
combustor of the 60 L - tank. The results suggested the possibility of the new method as a screenlng
test of gas - generating agents.