EXPLOSION OF TRI-NITRO-TOLUOL AND AMMONIUM NITRATE AT FACTORY NO.7, KENT.
Accident No. 110/1916.
No. CCXVII.
REPORT
To the
RIGHT HONOURABLE THE SECRETARY OF STATE FOR THE HOME DEPARTMENT
On the
CIRCUMSTANCES ATTENDING THE EXPLOSION, WHICH OCCURRED ON THE 2ND APRIL, 1916, AT THE
FACTORY OF THE EXPLOSIVES LOADING
COMPANY, LIMITED, AT UPLESS
MARSHES, FAVERSHAM, IN
THE COUNTY OF KENT
by
MAJOR A. COOPER-KEY, C.B.
H.M. Inspector of Explosives.
17th April, 1916.
Home office,
Whitehall, s.w.
17th April, 1916.
To the Secretary of State,
Home Office.
Sir,
I have the honour to report that in accordance with your Order dated April 3rd, 1916 I have held an Inquiry into the circumstances attending the explosion which occurred on Sunday 2nd instant at the factory of the Explosives Loading Company, at Uplees Marshes, Faversham.
By this explosion, the most disastrous in the history of my Department, no less than 106 persons were killed on the spot or died later from their injuries, and 97 were injured. Of those killed 20 were employees of the Cotton Powder Company and 4 belonged to the military guard.
Description of the factory.The factory at which the accident occurred was established in 1912 under an Amending Licence granted to the Cotton Powder Company whose works immediately adjoin. The original purpose for which the factory was erected was the compression of tri-nitro-toluol into charges for shells, torpedoes and mines, and until the outbreak of war no other explosive substance was in use at these works. Since that time, however, the management has been called on to fill shells and bombs with “amatol”, a mixture of tri-nitro-toluol and ammonium nitrate.
Circumstances of the Accident.At about 12.10 p.m. on Sunday, the 2nd instant, it was noticed by Mr. Underwood, Clerk of Works to the building contractors, that a heap of empty linen bags recently used for T.N.T. and resting against the matchboard wall of building No. 833 had caught fire and was burning at one corner. He at once gave the alarm at the office and Mr Palowker, the assistant manager who was in charge during the absence of Mr. George Evetts, the manager, at his mid-day meal, took immediate steps to deal with the matter by means of a hose and manual engine supplemented by fire-buckets. These failed however to have the desired effect and by the time Mr Evetts arrived the fire had obtained a good hold, and he had to decide whether to make an attempt to save a disaster by removing a number of cases of T.N.T. lying in the immediate vicinity of the building or to order everyone to retire to a safe distance and let the fire do its worst. After full consideration he decided that unless these cases were removed to a safe distance there was a considerable probability that the fire would spread to other buildings and that the consequences might lead to a disaster involving not only his own factory but also the cordite buildings belonging to the Cotton Powder Company, and that it was therefore his duty to prevent the fire spreading.
For over an hour water was poured on the fire without effect and at about 1.20 p.m. shortly after Mr. Evetts had given the order to retire the contents of the building detonated with appalling results to the crowd of men congregated around, very few escaping death of serious injury.
Damage to material.Simultaneously with the explosion of this building the two final washing and filtering houses belonging to the recently erected nitro-glycerine plant of the Cotton Powder Company situated at a distance of about 120 yards and each containing a considerable quantity of nitro-glycerine blew up in sympathy, and about twenty minutes later one of the buildings used for moulding “amatol” charges, and after another twenty minutes a fifth containing filled primers for naval mines also exploded after having been set on fire by debris projected by the previous explosions. Of these five buildings no trace was left, the crater formed at the site of the original explosion being no less than 50 yards in diameter and ten to fifteen feet deep. A plan prepared for me by the management, but which I have not thought it necessary to produce with this Report, shews very clearly that within a circle of 225 yards radius drawn from the site of the first explosion as centre every building of ordinary light construction was totally destroyed, whereas outside that circle no building was completely wrecked although many of them were so damaged as to be incapable of repair. In one of these shell-filling buildings erected prior to the war and situated within the fatal ring the reinforced concrete partitions were, however, not even cracked.
In spite of the fact that the buildings belonging to the Explosives Loading Company are well spaced at distances from 60 to 70 yards apart not more than three or four out of a total of about 30 escaped serious structural damage, but it must be noted that, with the exceptions mentioned above, all these buildings were lightly constructed of wood and were not screened by mounds.
In the adjoining works of the Cotton Powder Company the damage was widespread bit by no means regular. In addition to the six buildings within the 225 yard radius which were totally demolished not more than five were so damaged as to be incapable of repair and a further five seriously shaken but capable of repair. A very large number, however, amounting approximately to 150 to 200 received minor injuries or slight structural damage, a somewhat noteworthy feature being that the new magazines at the Harty Ferry extension in the extreme south-east corner of the factory area and therefore further away from the explosion centre than any other part of the works were more seriously damaged than the large majority of the intervening buildings. It has been suggested as a probable explanation that the greater apart of the factory is built on a floating crust above the marsh whereas these magazines were on practically the only solid ground in the area. As they were built right into the hill and well screened by mounds this explanation attributing the damage to the earth wave as distinct from the sound wave is very possibly correct.
So far as I can ascertain, although the effect of the explosions was felt many miles away, the damage outside the factory seems to have been limited to broken windows, falling ceilings, and removal of roofing slates, but here again, as is so usual in the case of a big explosion the wave seems to have been subject to no rue. Faversham, only four miles away, escaped almost entirely whereas Southend and Shoeburyness at fifteen to twenty miles suffered considerably, and I am informed that 2 windows were broken at Farlow in Essex fifty-two miles away.
Cause of the Accident.There is little to doubt in my mind that the original fire in the heap of empty bags was due to a spark from the neighbouring boiler house. The three flues of this were each fitted with a spark-catcher, but this at best is of doubtful efficiency and the wind was blowing almost directly from this house towards the heap of bags. It is possible of course that a workman had been smoking and had thrown a cigarette end of a glowing match on to the bags which, impregnated as they were with T.N.T. dust, would easily ignite, but having regard to the time at which the fire was first noticed, viz. just
before the dinner hour, and to the strict regulations against matches and smoking which had only recently been emphasized by prosecution and fine, it is extremely unlikely that any employee even if he had so far infringed the rules as to smoke on the premises would have failed to take the precaution of completely extinguishing his match or cigarette before throwing it away.
The theory of spontaneous ignition of the heap of bags has also been raised but from the description given by Mr Underwood of the first appearance of flame low down at one corner I consider this an improbable explanation, although having regard to the period of time, about three months, during which these bags had remained undisturbed it cannot be entirely disregarded. Lastly, it is not inconceivable that a mischievous person may have put a match to them on his way to the mess room, but in view of the fact that no one could have foreseen that the ignition of these bats would have had such serious consequences an unsupported surmise of this nature amounts at most to the vagest suspicion. The arrangements for preventing the entrance of unauthorized persons appeared to me to be very strictly carried out and in addition to the military guard 128 strong, no less than 24 civilian patrolmen are employed at the combined factories.
All things considered then I am, as already stated, firmly of the opinion that a park from one of the flues of the boiler house offers by far the most probable explanation of the fire. The boiler was situated at a distance of about 50 feet from the building against which the bags were lying, and contained three boilers, the flues of which were 37 feet high. The spark arresters were of a type consisting of a succession of baffle plates against which the flue-gases impinged and were fitted in consequence of the former frequency of visible sparks. In spite of these, however, sparking from the flues seems still to have occurred at times and only the night before the explosion two of the patrolmen reported that they had extinguished an incipient fire from this cause between the boiler house and the T.N.T. store.
Had the building contained T.N.T. only, it is probable that no explosion would have occurred but that the fused material would have spread over the ground and burnt away quietly. Unfortunately, however, owing to the alteration in the loading arrangement, to which references has already been made, this building had for some time been used as a store for Ammonium Nitrate and about 150 tons of this substance were present in it at the time. As the building was only licensed for T.N.T. the use of it for Ammonium Nitrate was, strictly speaking, a contravention of the terms of licence but, had the matter ended there, the substitution of a practically non-explosive ingredient for a compound which, although particularly insensitive to friction or shock undoubtedly possesses explosive properties would be regarded merely as a technical irregularity to be covered by an amendment of the licence at the first opportunity. But owing to the excessive quantity of T.N.T. delivered at the works the magazines specially erected for the storage of this material were all full and of the surplus about 15 tons were deposited in the ill-fated building and the remainder appears to have been lying on the open ground outside it. As therefore, the theoretically most efficient proportions of these ingredients are about 20 per cent of T.N.T. to 80 per cent of Ammonium Nitrate there were present in the building alone the makings of at least 75 tons of high explosive.
But although the management cannot be completely exonerated from blame for allowing this departure from the conditions of the licence, it must be remembered that at the present time rapidity of output is of the first importance, and from this point of view it is extremely difficult, if not impossible, strictly to adhere to the exact letter of a licence. Changes are called for almost daily and have to be met in the way that appears best at the time with due regard to the essential matter of output. This particular combination of ingredients even when intimately mixed to form Amatol, an explosive now commonly used for the bursting charges of shells and bombs, is regarded more as a fire risk than as an explosion risk, and it may well be that the responsible officials failed to appreciate the danger they were running in keeping these two ingredients in the same building or that appreciating the risk they considered it justified by the urgency of the national requirements.
That in moderate quantities a mixture of these two ingredients constitutes a fire-risk only would seem to be indicated by the length of time the fire was burning prior to the explosion. In fact according to the evidence of the Works Manager it had passed its maximum fierceness at the moment the explosion occurred.
