CA1087854A - Gas generating composition - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
A high burn rate nitrogen gas generating composition is provided which may be used as an igniter or booster in conjunction with slower burning mixtures. The composition comprises a stoichiometric mixture of an alkali metal azide or earth metal azide, an inorganic oxygen-supplying salt and a doped iron oxide containing in its particle lattice a small amount of at least one other metal oxide. The composition is characterized by a high burn rate, insensitivity to impact ignition or detonation, low ignition delay time and low output of toxic by-products.

Description

~7~ c I L 5 9 0 Thls invention relates to a chemical, pyrotechnic gas generating, composition in solid form capable, upon ig-nition, of rapidly producing large volumes of non-toxic gases. In particular the invention relates to a gas gener-ating composition which is characterized by the linear burn rates at 2000 psi adjustable in the range of 0.5 - 5.5 inches/
sec, relatively low flame temperature Tp ~1660K, and a non-detonability either in small quantity or in bulk form during manufacture and transportation. These characteristics make the composition of the inventlon particularly valuable as a booster for use in association with a main charge in the in-flation of safety crash bags in vehicle occupant passive re-straint systems.
The use of protective gas inflated bags to cushion vehicle occupants in a crash situation is now widely known ~ -and well documented. In the first devised systems of this type a quantity~of compressed, stored gas was employed to inflate a crash bag which upon inflation was lmposed between ~ -the occupant and the windshield, steering wheel and dashboard ;......... .:
of the vehicIe. In response to rapid deceleration of the vehicle, as in an accident situation, the stored gas was re- ;
leased through a quick-acting valve or the like to inflate the crash bag. Because of the bulk of the stored gas appara- `
tus, its generally slow reaction time and its maintenance difficulties, this stored, pressurized gas system has now ~ ~
largely been superseded by a system which utilizes the gases -;
generated by the ignition of a chemical gas generating or pyro-technic substance or composition. Such a chemical system employs a suitable sensing means and an ignition means such as an electrically activated squib or the like to ignite the ?~

,.. .
.. ~ .

- . . . . . .. ,. ... .. . . . ~, . .

7~

:, gas generating composition.
A large number of quick-burning, gas generating compositions have been proposed for crash bag inflation pur-poses, many of which have proven deficient in one respect or other. It has been a preoccupation of the industry to develop a gas generating composition which, without being detonable, combines the essential features of a short induction period, a burn rate which is rapid but even without any explosive effect, a high bulk density so that only small volumes of composition are required to produce large amounts of gas, the production of non-toxic gases so that vehicle occupants are not endangered in the event of a leak or during the vent-ing of the crash bag after deployment, the production of gases at a relatively low temperature so that damage to the crash bag is minimized and occupants are not burned, good filter-ability of the reaction products so that ash or hot solid . j .
, residue particles are simply removed from the gas stream and str~ong physical form so that long periods of storage can be a;ttained under wide ranging conditions o~ temperature and .,;. .
shock. :While some or other of these desirable properties arefound .in known chemical gas generating compositions, it has ~ not been possible to provide a single composition which satis .~ fies all the requirements of the industry since many of sought characteristics are mutually exclusive in a unitary or single ; composition. For example, there is a conflict bétween a de-sired rapid ignitability and a relatively slow burning. The , ~
I most common solution to this problem has been to use two dif-., . ferent compositions:- the first, or "booster" charge located at the ignition source, provides a rapid means of ignition of . 30 the adjacent second, "main charge" comprising a slower burning ,. .
~ 2 -, . ~ . . .
.: :. . . . . . . . ,. . ~ . . . .

85~L

gas generant. Depend'ng on the design of the generator, the booster may constitute up to 30% of the total composition.
Several potentially serious problems are associated with the use of booster charges. First, a high flame temper-ature leads to possible damage to the gas filtering means.
The compositions may be sensitive to impact and may be deton-able. The products of combustion may be toxic.
A novel solid, chemical gas generating composition has now been invented which is safe, efficient and durable and which satisfies substantially all of the requirements of an ideal gas restraint system. The composition of the inven-tion comprises a substantially stoichiometric mixture of an alkali metal or alkali earth metal azide and an inorganic oxidizer salt and from 1% to 31% by weight of iron oxide con-taining an adulterating amount of a second metal oxide within its crystal lattice. The composition is characterized by a short ignition delay time~ a rapid burn rate, high heat out-put, good filterability and low amount of toxic by-products.

.;
While gas generation or pyrotechnic compositions comprising two component mixtures of, for example, an alkali metal azide and an oxidizer such as potassium perchlor~te are known (see United States Patent No. 2,981,616) and while two component gas generating mixtures of an alkali metal azide and a metallic oxide are known (see United States Patent No.
3,741,585) it has now been surprisingly found that a three .. . .
component mixture of metallic azide, oxidizer salt and adult-erated or doped iron oxide provides improved burning proper-' ties which were both unpredictable from the prior art compo- ;
sitions and superior to the performance expected from blends of the aforementioned two component compositions. The im-... .

~ 3 ~

... .

~,785~

proved and uneXpected properties are of a synergistic nature The composition lends itself to use as a booster material to ignite less rapidly burning compositions used in association with it.
The azide used in the composition of the invention is preferably sodium azide or potassium azide since these materials are most easily procured in commercial quantities.
However, any alkali metal or earth ~etal azide may be em-ployed.
The doped iron oxide ingredient of the composition of the invention is of the type disclosed in applicant's co-pending application entitled "Adulterated Iron Oxide of High Chemical Activity" and filed concurrently herewith. Such an :
; iron oxide is characterized as containing not more than 1%
by weight of at least one other metal in the form of a metal oxide as an adulterant within its crystal lattice.
While a wide range of inorganic oxygen-supplying l salts may be employed in the composition of the invention, `
i those which are used by preference of availability, reactivity i 20 and stability are~sodium nitrate and potassium nitrate. Other ~suitable~oxidizer salts include barium perchlorate, potassium .i perchlorate and barium peroxide. The maximum quantity of any perchlorate used in the composition of the invention can be determined by detonability tests. The particularly preferred oxidizing salt of this invention is sodium nitrate.
' Since the;rate at which the gas generating compo-sition burns largely determines how quickly gas will be pro-duced to fill the crash bag, much attent.ion has been directed by the industry to the development~of a composition which has ~ a rapid burn rate yet not so rapid to produce an explosive ~ .

.

.

.:

~7~3~4 evolution of gases. It has been found that the burn rate of a given gas generating composition depends on a number of factors including the -temperature of the ignition means, the physical form of the mixture and the size of the par-ticles of the ingredients employed. The preferred composition of the present invention makes use of alkali metal azide or alkali earth metal azide having a grain size less than that held -on a lO0 Tyler mesh screen. The pre~erred physical form of the composition, unlike the milled powder or the pressed pellet of prior art compositions, is a water-dampened extru-date which has been dried to form an agglomerated granule.
A stoichiometric mixture of azide, together with an oxidizer salt and from 1% - 31% of the total weight o a doped iron oxide are combined with about 12% by weight of water and a small quantity of inert bentonite clay which acts as an ex-., ' .
trusion-aid and a binder. After mixing in a muller type mixer, the water-dampened material is extruded through a screw-.-.j, ..
type orming extruder equipped with a die. The size and shape ~ of the die opening will depend on the desired performance.

;~,!; . 20 Suitable forms include solid cylindrical extrudates of dia-..:
meters from 1.5 to 13 mm, hollow cylinders of outside/inside diameters from 3.5/l.0 to 7.5/2 mm, flat ribbons and star shapes. In general, the extrudates are cut into grains having a length-to-diameter ratio o 4:1. Following extrusion, the material is dried, and screened.
The agglomerated granule when made by the method of wet granulation extrusion and drying provides a homogeneous :"
mass highly resistant to breakage and separation of its con~
stituents during the shaking and impact imposed by the movement of an automotive vehicle. Additionally, the dried granule '' ~tl87~3S4 contains pores, volds, crevices and fissures which contri-bute to rapid burning. Furthermore, the absence of any carbonaceous binder material recluces the production of con-taminating gases.
The flame temperature of a pyrotechnic composition may be calculated accurately from the thermochemical para-meters of the system. The three component system of the present invention can be considered as a blend of two fuel/
oxidizer sub-systems - for example, stoichiometric NaN3/NaNO3 and NaN3/Fe2O3 mixtures. It can be shown that as long as the heat capacities of the reaction products of the two sub-systems are similar, the flame temperature of the system depends ~` linearly on the composition. Since the burn ratè depends pri-marily on the flame temperature by the Arrhenius type of dependence, it follows that a plot of a logarithm of burn rate for the mixture of the two .sub-systems should be also a linear~function.
A fuller understanding of the invention herein .:
described may be had by reference to the accompanying drawing which shows in graphical form some properties of various gas ; generating compositions.~

~ In the accompanying drawing the burn rate is plotted ,.j as a function of the composition of the system. The values on the left hand side of the graph refer to the stoichio-metric mixture of NaW3 with Fe2O3, and are labelled A. The , ~ values on the right hand side of the graph refer to the stoi-chiometric blend of NaN3 with NaNO3, and are labelled B. The ~ ~ two solid lines show experimental results. The lower line, `~ No. 1, represents the data obtained using commercial Fe2O
(see Example Nos. 12 - 21 below). The upper line, No.

., . .
.~ ~

s~ :

represents data obtained when nickel doped iron oxides were used (see Example Nos. 22 - 32). The broken lines represent the dependence of burn rate on composition which are expected for the blends of A and B sub-systems in the absence of any catalytic or any synergistic effect. In the drawing, the i burn rate of the stoichiometric blend of KC104 with NaN3 is also indicated - this blend pre~iously used as a standard ` booster in air bag systems serves as a measure of the relative rapidity of burning of the cooler NaN3/NaNO3/doped Fe2O
system.
EXAMPLEs 1 - 10 A series of tests were undertaken to determine the sensitivity (detonability) of typical two-component and three-~ ~ , ; component gas generating compositions. In these, stoichio-metric compositions of sodium azide and various inorganic oxidizers were prepared and tested. The tests were performed ~
in two stages. In the first stage 40 g of material was placed in a steel pipe of the inside and outside diameters respect-ively 18 mm, and 26 mm, and initiation~attempted with a No. 8 electric blasting cap. If initiation was negatlve or doubi-, ~, ;; ous~ the material was tested in the second stage. Here 5 kg of material was placed in a steel pipe of I.D. ~ 10 cm, O.D.
= 12 cm and initiated with 320 g of TNT/PETN primer. The steel - pipe was insulated with 1" thick layer of polyurethane foam.
The material in the pipe was conditioned in a hot chamber for - 24 hours at 40C, then tested under 7 m of water. The temper-ature of the material and continuous velocity of detonation were monitored. ~ -The gas generating material for the tests was pre-pared by dry-blending sodium azide with oxidizer(s), both , .

lOB7854 finer than lOO Tyler mesh screen size, followed by wet granu-lation, drying and screening. Usually two grain sizes were used: 4 - 14 mesh and 14 - 35 mesh.
The results of these tests are summarized in Table I.
TABLE I
. _ Example Oxider(s) and Results No. their molar ratios . .....
. 1 KC104 detonates

2 KC104:SiO2 = 4:1 1. ;~

3 KC104 :Nio ~ 4:1 partial detonation

4 ~iO non detonating S NiO:KC104 = 4:1 . 6 Fe23 " ..
7 NaN03 ll ll ., 8 NaN03:Fe203 = 9:1 . " "
' 9 " " - 1:1 .. ..
l 10 " " = 1:9 ". 17 :' .. ._ ._ .__.__ ., .. ~ i : ' The results shown in Table I demonstrate the poten-tial detonation hazard present in some conventional or typical j~ gas generating compositions. All the exceptionally fast burn-:
ing compositions (Examples 1 - 3) were shown to be detonable and hence unsuitable for use in automobile safety crash bag systems. -., . ;ExAMpLEs 11 - 21 :~ .
.. l In order to demonstrate the catalytic effect of com-;, mercial grade iron oxide on the burning characteristics of gas : generating compositions, mixtures were prepared by the method ~ of Examples l - lO comprising stoichiometric mixtures of so-i~ 30 dium azide and sodium~nitrate together with varying amounts : 8 i'i, ~.

85~L
of commercial grade iron oxide. After preparation of gran~
ules these were crushed and -28 mesh material was pressed into 20 g, 25.4 mm diameter pellets using 680 bars of hydro-static pressure. The pellets were inserted, one at a time, into a small specially designed holder, placed in 185 ml pressure vessel and ignited with 2.2 g of gasless (Si t PbO2) mixture. Pressure (P) vs. time ~t) and (dP/dt) vs. t were recorded from which the ignition delay time, ~ and a reduced burn rate R , (l/P) (dP/dt)maX were calculated. The data are presented below in Table II. The numbers represent an average value of 3 - 5 separate tests. For comparison, the results for composition containing XC104 (Example 11) are also given; the burn rate of this material was previously recognized to be adequate for use in the inflatable safety system.
-- The commercial-grade iron oxide used has a purity of at least 98.6%, an average particle size 0.4 (~m) (defined as length of acicular crystals), a specific surface area 9.5 (m2/g) and a bulk density 0.9 (g/cm3). ~ ~;
.
:

: ;' .' .

~'' : , ~:
.'' , : ~ ;

.~ -- g --. . --.

,.i : i.

1~7~5~

TABLE
_ . . .. . __ .. .. . ~ _ . . . .
¦ Example Composition ~wt %) Burn Parameters No _ ~aN3 Fe23 NaNO3 Ignition ~1ax. IBurn delaY time Pressure rate Na*
_ __ ~ ~ I
(ms) (~si) (s~l) __ l _ 11 58.96 (KC104 = 41.04) 150 4340jl5.3 _ 12 61.9538.05 - - _ 670 ' 1455~ 2.1 3 13 65.4230.44 4.14 275 1881~ 5.0 5 - 14 68.8822.83 8.29 190 ~340~ 6.8 6 72.3415.22 12.44 175 26901 9.1 7 16 75.81 7.61 16.58 156 312910.1 8 17 7~.95 5.81 18.24 140 320510.2 7 , 18 77.72 3.42 18.86 141 327810.7 8 19 78.23 2.28 19.49 144 332010.5 7 ,, 20 78.75 1.14 20.11 138 329110.8 8 ~ ,21 79.27 20.73 162 346910.7 ,8 `~! *Na,represents the relative content of free sodium in cinder, '',, - and is evaluated on the arbitrary scale 0 - 10.
:, ~, From the results in Table II it can be seen that for ' the last four compositions the burn parameters are nearly con-., stant, indicating that indeed a catalytic effect of Fe2O3 com-, pensates the faster burning composition o azide and sodium '~ nitrate with the slower burning composition of azide and iron oxide. The catal~tic effect is however inadequate to bring the burn parameter of the three component compositions to the level ~ , .. . .
of the composition containing KC104 (Example No. 11). It is worth noting that the flame temperature of the perchlorate com-'~! position Tp = 2223K is significantly higher than that of any ,1 ~
other formulae listed in Table II (Tp s 1932K).

~, 30 Samples were prepared by the method described in ,1 ' ~, .
~, - 10 .
`:. . . :

~L0~7~5i~1L

Examples 11 - 21, and tested in an analogous manner. How-ever in this case commercial iron oxides were replaced by an iron oxide containing a small amount, less than 1% by weight, of a nickel dopant. In Table III below the results of these tests are recorded. The composition of the samples ; was kept constant: NaN3 = 77.53, Fe2O3 _ 3.81 and NaNO
~ 18.66 wt%.

; TA~LE III ~-- Iron Oxide Burn Performance Ex. Ni Particle Sp. Sur- Ignition Max. Burn ~
No. contentSize face area delay time Pressure rate Na -(%) (um) (m2jg) (~s) tp~i) (s-l) 22 0.044 0.54 20.8 112 3343 12.5 _ 23 0.18 1.10 110 109 2610 1~.5 7 ~-~i 24 0.50 3.50 87 102 32~4 15.4 9 0.05 ~0.70 95 100 2972 15.0 6 26 0.045 0.60 62 99 3028 16.4 6 27 0.045 0.58 37 117 3000 16.2 8 . 28 0.045 0.59 9 106 3438 14.2 7 ~`i`;~ 29~ 0.045 0.51 14 125 3031 15.5 8 0.045 0.43 11 115 3210 15.1 7 ~` ~ 31 0.045 0.53 12 85 3288 15.2 8 32 0.~045 _ 0.40 _ 4 103 __ 3151 15.4 7 The results shown in Table III demonstrate improved burn ,, .';1 parameters comparçd with the compositions described in Examples 12 - 21 and in 1'able II. The improvements may be directly at-tributed to the use of a nickel doped iron oxide ingredient.
I EXAMPLES 33 - 36 ~-'l Samples containing various amounts of nickel doped iron ~ ~

oxide were prepared by the granulation pro~ess described in ~ ;
Examples 1 - 10. The nickel doped iron oxide, evaluated pre~
.
... j . ~
`-c', - 11 '- " ~ ~

.

, .
~, 78~
viously in Example 24 was used. Granular material was loaded (12 g) into a small generat~r and ignited with S-140 military type squib. The evaluation of the burn performance was identi-cal to that described in Examples 11 - 21. The results are listed in Table IV, along with the data for com~osition of ~;.
Example No. 11 containing KClO~.
TA~LE IV

Ex.¦ Wt. % Compo~ ;ition Grain Burn Performa: lce No. NaN3 Fe23 NaN03 Mesh Ignition Max. Burn Size delay time pressure rate .. ... ~ . _ . . (ms) (psi) _ (5~

33 79.27- 0.00 20.738.-14 16 2952 283 . . 14-20 15 3283 398 .. 34 77.53 3.81 18.668-14 }5 2827 403 . 14-20 1.1 2938 580 .. 28-35 7 2804 850 . 35 61.95 38.05 0.00 8-14 183 764 35 .; _ _ . . 14-20 64 785 45 36 NaN3 t KC104 8-14 22 2982 402 ~- a in Ex. .14-20 10. 2960 566 : No. 11 20-28 11 2930 640 . .. 28-35 12 2830 620 .
From the results in Table IV it can be seen that for the same grain size, the composition containing NaN3 + NaNO3 (Ex.
. 20 No.. 33) burns slower than that containing NaN3 + RC104 ~Ex. No.
36); however the three-component composition NaN3.+ doped Fe2O3 +

~ NaNO3 tEx. No. 34) ignites and burns the fastest. In the ab-.j.
. sence of the catalytic effect o~ Ni-doped Fe2O3 the burn rate .' . of the mixture would be expected to be-at an intermediate rate ''! determined by the amount and the burn rate of the-two blends , containing NaN3 + NaN03 and that of NaN3 + Fe203 (see Example Nos. 33 and 35 respectively) ~258 and 363 (5-1) for 8 - 14 and , 14 - 20 mesh respectively). It can be seen that the actual burn ~.
rates of the three component mixtuess are 56 - 60% higher, indi- -~
:~ 30 cating a strong synergistic effect.
'~ , ' : ,', `, - 12 - ~ ~
' ~ .
.. . .

85~

Samples containing varying amounts of nickel doped iron oxide were prepared. The ingredients were combined to-gether by mulling for 90 minutes with 14~ of water, then ex-truding the resultant paste through a die plate with 1.6 mm holes (length to diameter ratio 4:1), drying the grains and screening. The samples in form of solid cylinders (about 3 mm ~ }ong and 1.6 mm in diameter) were tested as in Examples 33 - 36.
- The high-torque, high-energy commercial mulling ~-equipment used in preparation of these samples precluded a manufacture of the detonable compositions containing KC104 in the same manner. However, on the basis of experiments with other compositions it can be estimated *hat the high density ` extrudates burn with the rate approximately l/2 of that measured for 8 - 14 mesh hand-granulated material~ The result of these ' tests are presented in Table V.
TABLE V

Ex. Wt.~% compositiohBurn Performance No. Ignition Max.~ Burn ~ '-NaN3 Fe23 NaN03delay time pressure rate ~i . ' ~_ ' ~ r ~ (mS) (psi) (s~l) 3779.27 _ 20.73 84 _ 156 20 38 77.53 3.81 18.66 24 2922 222 3972.14 14.39 13.47 3~ 2056 176 ~069.01 21.66 9.33 41 1698 139 4}65.87 28.95 5.18 43 1582 112 4263.83 33.68 2.4~ 64 1338 82 ;~ 43 61.95 38.05 _ 422 1084 14 ,. ... .. .. , .. , _ _ : - .
-' The xesults of Table V show that the composition of ~j~
1 ~ Example 38 burns-significantly faster than the NaN3 + NaN03 I sample of Example 37 and with a rate at least similax to that .
which can be expected from N~N3 + KC104 formula.

30It can be noted that, depending on the desired per-: ........................................................................ ..
, ~.

.~ -- 13 --.. . .

lal~P7'8~

formance, the bur~ rate of the material can be easily adjusted.
In all cases the flame temperatures of the compositions of this invention were lower than those of the previously used compositions. This lower flame temperature is essential for the reliable performance of gas generators equipped with fine filters. I~ the flame temperature of the material is too ` high, pinholing of the filter pack occurs, leading to release of very dusty gas. The flame temperature of the composition containing 10% of NaN3/Fe203 and 90% of NaW3~NaN03 blends was ; lO computed as Tp 2 L641K, that containing 75% of NaN3/Fe203 and 25% NaN3/NaN03 blends as Tp = 1405K.

The samples used in Example Nos. 38 and 40 were used in standard automobile air-bag systems as boosters in the commercial passenger side and steering wheel generators respect-ively, as replacements for the standard booster compositions . .
containing KC104. In both tests an excellent performance resulted which was superior in reliability and gas purity to that of the standard composition.

.
..
", . .
, .

. ~
;
AGENT FOR APP~ICANT
DONALD G. BALLANTYNE

:'~ ' ' .

: : :
,:
.. '' ' ' .
i . :
. 14 .... ~

Claims (3)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A solid, nitrogen gas generating composition comprising a substantially stoichiometric admixture of an alkali metal azide or alkali earth metal azide, an inorganic oxygen-supplying salf and from 1-31% by weight of the total composition of a doped iron oxide, said doped iron oxide comprising an acicular iron oxide particle having an amount of up to 1% by weight of another metal oxide impregnated into the iron oxide structure.

2. A composition as claimed in Claim 1 wherein the azide is selected from sodium and potassium azide.

3. A composition as claimed in Claim 1 wherein the oxygen salt is selected from sodium nitrate, potassium nitrate, barium perchlorate, potassium perchlorate and barium oxide.