712

THE COLLIERY GUARDIAN.

April 4, 1913.

any rate suggests two other objections acting in the
opposite direction, the first being that the great mass of

metal conducts the heat away rapidly from the pro-
ducts of combustion, and the second is that this factor
increases with increase in the factor of danger as, with
an explosive which will pass only with a charge limite of
8 oz., the cartridge occupies only the bottom 7 or 8

inches of the bore and has over 3 ft. of cold metal further experiments are needed to establish this point,
surface to cool the products before they reach the test as extended experience will probably show that the gases
mixtures, whilst an explosive that has a big charge occluded in the coaldust, especially oxygen, play an
limite nearly fills the cannon, and the top of the charge important part in the temperature at which ignition
and detonator fire direct into the mixture without the i takes place, and, unfortunately, the dusts tested at
cooling help of the metallic contact, so that, to a ! Eskmeals were all dried at 107 degs. Cent, for an hour

certain extent, the dangerous explosive is favoured and
the safe one penalised.

We will now pass to the test mixtures, which consist
of :—

(1)	A mixture of 55 cubic feet of Rotherham coal-
gas with 299 cubic feet of air, this filling the 354
cubic feet of space at the end of the gallery into
which the shot is fired, and being retained there by a
paper diaphragm; this quantity is equivalent to
15’5 per cent, of gas.

(2)	120 oz. of coaldust, 40 oz. of which are
arranged on a board in the first 10 ft. in front of the
mouth of the cannon, whilst the remaining 80 oz. are
placed in heaps in the remaining 40 ft., so that when
the shot is fired the rush of the products of com-
bustion stirs up the dust cloud.

When the testing gallery was removed from Woolwich
it was generally supposed that the Midlands had been
fixed upon as yielding a supply of pit-gas for the test,
but the authorities came to the conclusion that the
supply of coal-gas would be of more even composition
than a natural gas, and that as there was only some
50 degs. Cent, difference between their recorded ignition
point it was better to adopt coal-gas for making the
mixture. If the Rotherham test is accepted as a purely
empirical test intended to place explosives in their order
of danger this conclusion is justified, but if there was
any intention of making the tests in any way resemble
practical conditions it was a mistake. The French
Explosives Committee many years ago adopted 650 degs.
Cent, as the ignition temperature of pit-gas, but found
that this temperature had to be applied for 10 seconds
before the gas ignited, whilst for its immediate ignition
a far higher temperature had to be applied. It is this
which causes the danger with gunpowder and other slow-
burning explosives, which, in the event of a blown-out
shot, would give a flame of long duration, whilst with
more rapid explosives, which on detonation give a
flash lasting only for the fraction of a second,
the flash or the products have to come in
contact with the firedamp whilst still at a very high
temperature to cause ignition. With Rotherham coal-gas,
however, which contains approximately half its volume
of hydrogen, the temperature would have to be con-
siderably less than that needed for firedamp to give
immediate ignition, so that a flash from an explosive
having a temperature far too low to ignite a pit-gas
mixture would easily ignite the coal-gas mixture.

In making the dust and air mixture, its sensitiveness
will depend upon (1) the ease of ignition of the dust, the
degree of fineness to which it is reduced, and the other
factors which have been studied so fully at Al tofts
and Eskmeals, one of the most important being its
freshness; (2) upon the percentage of dust to air at the
moment the shot is fired into it; and (3) the percentage
of oxygen in the air. Through the courtesy of Major
Cooper Key, I obtained a sample of the dust used at
Rotherham, which is made by grinding coal from the
7 ft. seam of the Birchen wood Colliery, North Stafford-
shire, to a powder that will pass a 150-mesh sieve. On
analysis, the dust gave :—

Per cent.

Moisture .................... 2*40

Volatile matter ............. 29	60

Fixed carbon................ 65’41

Ash........................... 3	94

whilst its ignition point was 400 degs. Cent. It is
evident, therefore, that if the coaldust be suspended in
pure air with a percentage of 20 9 of oxygen, as soon as
the temperature is raised for a sufficient length of time
to 444 degs. Cent, the dust will ignite. In the second
report of the Explosions in Mines Committee, the
“ relative ignition points of a long series of coaldusts,
as determined at Eskmeals, are given, and out of 50
dusts two only are credited with an ignition point
below 1,000 degs. Cent, (i.e., 950 degs. Cent.), and four
are above 1,400 degs. Cent. In this series the sample
N 216 has much the same proximate composition as the
Rotherham dust, and is credited with a “ relative ”
ignition point of 1,065 degs. Cent. An examination of
the method employed shows that what has been
determined is the relative ease of ignition under certain

circumstances, and has nothing to do with the ignition
point.

The Eskmeals experiments, however, are of great
interest from another point of view. It appears probable
that coaldust, although ignited by a heat of 400 degs.
Cent, continued for a few seconds, requires a temperature
of over 1,000 degs. Cent, for immediate ignition, but

before being experimented with, which would expel test was added of also firing direct into coal-dust

these or, in any case, alter their nature.	( suspended in air, with the extra proviso that the

We next come to the influence of the percentage of charge limite should be at least equal in strength to a
dust in the test mixture. At both Rotherham and charge of 175 grammes of No. 1 Dynamite. It was then

Frameries the percentage of weight of dust to air in the found that of the 30 permitted explosives on the old
gallery is approximately 10, but as the dust is raised by list, five inflamed the dust mixture with a smaller

the explosive wave the percentage present in the path

of the outrush of heated products is probably far reduction in charge limite brought two of them below

higher, and will vary probably with the kind of the strength needed to pass the test. The three

explosion. It is felt, however, that having the dust at explosives which had their charge limite reduced but
rest resembles more nearly the conditions existing in a ' were still strong enough to pass the test were Yonckite
mine, where it is mostly raised by the explosion itself. I No. 10, Minolite Antigrisouteuse and Ammon Carbonit,
At every testing station it has been recognised that whilst by elimination from the old list it seems probable
the apparatus had its good and bad days, and that there j that Densite III. and Wallonite III. were the two
were undoubtedly meterological conditions which affected that had their charge limite so reduced as to fall below
seriously the results of the testings, but what the inter-. the necessary standard of strength. The reduction of
fering factors were has not ever been satisfactorily | the gas charge limite when the same explosives came to

explained. Early this year some experimental shots were
fired at Rotherham with two explosives, which proved
thoroughly satisfactory, one failing to ignite gas or dust
with a 32 oz. charge, the other proving safe with 30 oz.,
and as they were both stronger than most explosives
that showed any promise of passing, the manufacturer

was naturally elated, but decided wisely to have some
more experimental shots fired with cartridges from the
same batch before sending in the explosives for the
official test. Both mornings were raw and cold, the
explosives were the same batch, the dust freshly ground
in each case, and the tests carried out under identical
conditions, yet both explosives in the second test gave
fierce ignition with small charges in dust, although
the results in the gas mixture were the same as
before. The only difference between the two sets of
experiments was that the first, when the explosives
would have passed with the highest maximum charge,
were carried out when there was an abnormally low
pressure, the barometer standing at 29 3 in., whilst the
second set, in which both failed, were made with the
barometer standing at 30’2. The difference of an inch
in the barometric column makes a difference of about

3	per cent, in the weight of the oxygen present in a cubic
foot of air, and this is probably the factor that made
the difference between passing the test with a 32 oz.
maximum charge and firing the coaldust mixture with
a small charge. I know that some authorities consider
it is the volume and not the weight of oxygen in air that
counts, but it seems to me that it must be the number
of molecular contacts in a combining mixture that
governs the rate and ease of combination, and if experi-
ence proves that pressure has this effect, what becomes
of our tested safety when the explosives are used at the
pressures existing in a deep mine ?

The apparatus at Woolwich was altogether too small
to obtain any definite information as to the alteration in
“ maximum charge ” likely to be brought about by doing
away with tamping, but some experiments have been
made upon this point at Frameries which go to show
that the wetter dynamites suffer the least, having the
charge limite reduced only by one-half, whilst explosives
containing mixtures of nitrated aromatic hydrocarbons
and oxidising compounds suffer most, and have their
charge limite reduced to one-tenth of what it would be
with even a 4 in. tamping. It is clear, therefore, that
most of the explosives onjthe old Permitted List will fail
to pass the new tests with a “ maximum charge ” of
even 8 oz., whilst the Rotherham conditions of test are
so sensitive that many of the S.G-.P. explosives on the
Belgian list will also fail. Reading through the
percentage prescriptions for the explosives on the
Permitted List, it is apparent that one large class of
safety explosives has been arrived at by cooling down
the explosion of nitroglycerine by adding cellulose in
various forms and in sufficient quantity to ensure the
formation of products of incomplete combustion, and
then adding diluents of a character likely to stifle flame ;
whilst a second class has been formed by the reverse
process of “ waking up ” ammonium nitrate by adding

4	or 5 per cent, of nitroglycerine or trinitrotoluene to it
to ensure complete detonation, and then further cooling
by replacing some of the ammonium nitrate by sodium
or potassium nitrate with or without other diluents.
A third class consists of trinitrotoluene or other nitro-

substitution products, with oxidising salts, generally
nitrates of potassium, sodium and ammonium, and

diluents.

The marked characteristic that is to be found in the
products of combustion evolved by the first class is that
they contain large volumes of carbon monoxide, whilst
the products from the second class and from some
explosives in the third class contain free oxygen, often in
inconsiderable quantities. Already it has been recognised
at Rotherham that all explosives do not behave alike in
both gas and dust mixtures, some lighting the dust
more easily than the gas, whilst others show a tendency
to do the reverse. At Frameries, about 1905, the extra

charge than that needed to ignite the gas, and that this

be fired into dust was :—

Charge limite

In gas. In dust. Reduction.

	Grammes. Grammes. Grammes.			
Yonckite No. 10			800 .,	.. 500 .	.. 300
Minolite Antigrisouteuse		650 ..	. 400 ..	.. 250
Ammon Carbonit ..		400 ..	,. 300 ..	.. 100

On calculating out the ratio of combustible to oxygen
i set free for all the explosives in the present Belgian
Permitted List, these three are the only ones that evolve
over 15 grammes of surplus oxygen from the explosion
of 100 grammes of explosive, whilst calculation
shows that Wallonite III. and Densite III. yield over
20 grammes.

It is quite clear that the liberation of considerable
quantities of oxygen by the explosive must make the
dust more sensitive, and this is probably the chief
reason why the charge limite of such explosives is
reduced when fired into coaldust. As far (as I have
been able to trace this action, it is only when the
explosive liberates more than 15 grammes of uncom-
bined oxygen per 100 grammes fired that the charge
limite is affected, and there are explosives on the
Belgian list which give 8, 10, or even 12 grammes per
100 grammes fired, and yet have the same charge limite
in both gas and dust, and the reason of this is that with
these lower quantities the percentage of oxygen in the
products of combustion is well below the percentage
present in the air, so that the products render the dust
less sensitive in the test rather than more. I think this
is where one of the real practical dangers of the new
procedure is to be found: if any explosive contains an
excess of oxidising material, unless it yields over 17’5
per cent, of free oxygen in the products of combustion,
it would not show any effect in the test, but in practice
it would combine with the coal, and there would always
be the danger of sparks or even ignition of the coal.

I am quite aware that explosives giving a large excess
of free oxygen have been in use for a considerable time,
and that, as far as I know, accidents have not been traced
to them ; but it must be remembered that one great dif-
ference between the Belgian practice and ours is that no
shots in the coal itself are allowed except by special per-
mission, which is difficult to obtain.

Twenty years ago, in a course of Cantor Lectures
before the Royal Society of Arts, I pointed out the
danger of large volumes of an easily-united combus-
tible gas like carbon monoxide being generated by
explosives intended for mining work, not only because of
the poisonous properties of the gas, but also because a
very small percentage of it rendered air containing coal-
dust far more explosive than when no gas was present,
so that two shots fired in rapid succession might lead
to disaster. At the present time, however, the
use of nitroglycerine in a safety explosive to
the extent of more than 3 or 4 per cent, (a quantity
used in some explosives containing large amounts of
ammonium nitrate to ensure complete detonation) can
be accomplished safely only by the admixture of carbo-
naceous materials, such as wood meal and various kinds
of flour, to cool the explosion by forming carbon
monoxide instead of the dioxide. The larger amount
of dilution of the original explosives of this class needed
to pass the Rotherham test will increase rather than
decrease the volume of carbon monoxide given by the
watered-down explosive, and this accentuates the danger