THE DESTRUCTION OF THE LIBERTE.
By Lieut.-Commander Ralph Earle, U.S. Navy
The battleship Liberte, 14,868 tons displacement, forming one of the units of the second battleship squadron under the command of Vice-Admiral Bellue, anchored in the harbor of Toulon, was destroyed by fire and explosion on Monday morning, September 25th, 1911.
The weather was beautiful, and the ship had just started to swing to a light breeze on starboard bow. Soon after reveille had sounded, and while the crew were scrubbing clothes on deck, a noise, as of escaping steam, was heard together with a heavy sound or rumbling. This was at 5.17, and yellow acrid smoke with a powder smell issuing by way of the ammunition hoists and ventilation pipes from the starboard forward 7".6 magazines was quickly followed by blue and then yellow jets of flame. These flames with thick and suffocating smoke filled the ship rapidly. The crew went to fire quarters quickly, Lieutenant Bignon in charge, and unsuccessful efforts were made to flood the magazines. The men endeavoring to do this were driven from forward by this heavy yellow acrid smoke. Possibly also the intense heat of the fire had bent the operating gears of the flood valves.
The chief engineer, Lestin, in obeying a second order to flood the forward magazines, saying, "I've already tried," started forward again but never returned. About 5.27 to 5.35—time variously estimated within these limits—an explosion occurred under the foremast sending up a thick column of smoke followed by flames; three other explosions followed at intervals, the first putting out all the lights on board, thus adding to the difficulty of controlling events. Assistance was signalled for and sent promptly by the other vessels of the squadron.
Lieutenant Gamier took charge about this time, and, the flames seemingly subsiding, countermanded an order of Lieutenant Bignon to flood the central and after magazines, until he could learn more definitely the exact condition of affairs, and to quiet the growing panic.
The interior of the hull was burning, and from time to time jets of flame shot out of the foremast, these being caused by the explosion of boxes of 9 and 3 pounder ammunition.
The call to quarters was sounded, and the crew went to their stations bravely, and even those who had jumped overboard in the first panic endeavored to reboard the ship to return to their stations. Water could not be obtained in sufficient quantity in fire hoses, the attempts to flood forward magazines were failures, the whole inside of the ship was a furnace in which were burning all sorts of combustible material, bags, hammocks, linoleum, etc., and a thick smoke penetrated everywhere. The order was given to flood the amidship magazines, but the smoke and darkness rendered this impossible of accomplishment. The temperature between decks rose rapidly. The order at last was given to abandon ship. Then at 5.53, i.e., just twenty minutes after the first explosion, came the final and tremendous explosion which converted the Liberte into a mass of shapeless and half-sunken debris. The roadstead was covered with wreckage and the maelstrom cause by the explosion swallowed up many boats sent by other vessels to the rescue. This explosion was heard for over thirty miles.
From the testimony of Says, an artificer, who escaped miraculously, we learn that at the moment of this explosion, Lestin. and master machinists Champilly and Steiner, the latter wounded and covered with blood, refusing to save themselves, were on the berth deck struggling with the locks of the flood valves to the amidship magazines. Surely this was heroism of the highest type.
Debris was hurled in all directions, causing damage and death to other vessels of the squadron and to the rescuers.
The Republique was struck by a mass of armour, weighing several tons, about fifty feet forward of the stern, her plates were stove in and ripped off for a distance of twenty to thirty feet, an; half-way down to the water line. The Liberty's forebridge and two 12" shells landed also on the Republique. The damage to her was so great that she was docked as soon as possible.
The Democratic was the next heaviest sufferer. In addition to the above two vessels, the following sustained damage and loss of life: Verite, Justice, Jules Ferry, Michelet, Foudre, St. Louis, Carnot, Marsellaise and Edgar Quinet.
The final official figures as to the loss of life on the Liberte and the above mentioned vessels are:
204 killed or missing
136 wounded, and
48 slightly injured.
The decks, including two f.6 turrets, of the Liberte from abaft her forward boiler rooms and turrets were blown upward, backward, and doubled up on the quarter-deck of the vessel. The forward part containing the two forward funnels has disappeared. The vessel is now simply a mass of twisted and entangled plates, and together with detached fragments renders navigation in the roadstead of Toulon very hazardous.
Cause of the Disaster
A committee, headed by Rear-Admiral Gaschard, made a thorough inquiry, and, despite the ugly rumors as to socialism and malevolence have reported as follows:
- That there was no trace of malevolence.
- That the supposition that a fire either in magazines or in adjacent parts of the ship started decomposition of powder is disproved; and that
- The spontaneous ignition of a charge in the forward starboard upper 7".6 magazine was the source of the disaster; and
- No blame is to be attached to officers or crew, all regulations having been complied with.
The conclusions of this committee in general are:
- The burning of the powder, which twenty minutes later exploded the shells, and transformed the Liberte into a terrible wreck, started in a 7".6 magazine and not in the 9 or 3 pounder magazine.
- That in this 7".6 magazine there were no exercise, i.e., target practice, powders, all such having been expended at the previous target practice. The gunnery officer of the Liberte is very positive on this point.
- That the oldest B powder in this magazine dated only 1906, and that the other lots of B powder were of dates, 1907, 1908, 1909 and 1910. All these lots carried the same mark, BM15AM8 (amyl alcohol as solvent, 8%).
- That, therefore, the oldest B powder to which the explosion can be attributed was aged but five years; and
- That, by deduction, powders of the B type, five or more years old, can spontaneously ignite.
The powder in question was made at the Pont de Buis factory in 1906. As this lot of powder was looked upon as absolutely reliable, considerable uneasiness has naturally been caused in France by these conclusions. The fact that use of amyl alcohol in manufacture has been abandoned recently leaves a possibility that the powder made in recent years and now in service, will not thus prove unreliable. That its use was deleterious, however, is by no means a point that is agreed upon by chemists.
The destruction of the Liberte was thus started by the spontaneous decomposition of a charge of 7".6 ammunition. This set fire probably to the woodwork of the magazine, and the heat caused other charges to explode, not as a detonation, but as a blast furnace or escaping steam, thus adding to the heat in the magazine, and finally causing the melinite shells to burst. The bursting of shells loaded with melinite caused the destruction of the vessel. In the forward magazine there were 142 semi-armor piercing and 101 armor piercing 7".6 melinite shells. In the amidship magazine there were 388 melinite and 104 black a. p., and s. a. p. 7".6 shells. There is some doubt as to whether the shells of the forward magazine exploded, as the damage done to the ship seemingly indicates that the heavy explosion which destroyed the ship occurred in the amidship magazine.
Results of the Disaster
M. Delcasse, Minister of Marine, has ordered that, hereafter, 4 years is to be considered the limit of age for B powders considered suitable for aboard ship, and this limitation will likely be extended to magazines on shore. This is an expensive step, but, in the uncertainty which at present exists, it seems a wise one.
The monopoly held by "Les Poudres et Salpetres" has been broken and placed under control.
Naval officers will, in future, control and inspect the manufacture of powder.
The capacity of the factories of "Les Poudres et Salpetres" has been ordered largely increased, and is now turning out 3000 tons of B powder with diplenylamine used as a stabilizer in place of the now-apparently discredited amyl alcohol.
The factories controlled by the Bureau of Poudres et Salpetres were held to be guilty of:
- Bad workmanship, particularly as to the purity of the raw materials used, the purification of its nitrocellulose, and thoroughness of carrying out the various steps in manufacture.
- Slow and costly manufacture; some fifty-five cents per pound avoirdupois.
- Deceiving its customers; i.e., the naval officers.
- Refusing to experiment with and improve its product.
- Sticking to one standard routine.
As a result of the Iena explosion, black and brown powders were separately stored from B powders; additional inspections were provided also. The temperatures permissible for magazines were gradually reduced from 400 C. (104° F.), to 350 C. (95° F.), then 320 C. (89.6°F.), and now .25° C. (77° F.) is to be the limit. This does not appear to be the result of any sound reasoning or experiments however. During recent naval maneuvers temperatures over 30° C. or 86° F. were continually registered, showing the comparative inefficiency of the isolating system between magazines adopted for the Patrie class after the Iena disaster.
The cooling and flooding appliances on board all fighting vessels have been minutely investigated, and tested and practical means to improve them are to be adopted without delay. A few constructors are to be sent to the English fleet to study methods there. It is said that the arrangements for flooding magazines were so unpractical that, even were they in good order on the Liberte, they would have been powerless to save that ship in case of spontaneous decomposition. They are called "a challenge to common sense" by naval officers. In case of St. Louis class of French battleships, 30 minutes are required to flood the after 12" magazines. The water is brought to the lower level and not to the top of the magazines.
A committee of eminent specialists has been formed to examine what practical steps it is desirable to take in order to improve the stability and quality of the B powder, and to consider the question of modifying its composition.
General Notes on the Subject
Attention is called by many to the similarity in the phenomenon of the Iena and Liberte explosions. To quote liberally from the Vie Maritime.
"Thus it was on the Iena following the small explosions of the cartridges, as they took fire one after another, came dark and thick smoke with intense heat forming an unquenchable fire and twenty minutes later, the walls of the magazine having become red hot, the shells exploded. This was reproduced in the experiment at Gavres. Twenty minutes after the cartridges of B powder had been lighted, the shells started to explode. The ordnance experts witnessing this could not repress their astonishment.
And now this is exactly what happened on the Liberte, the itch starting spontaneously in a 7".6 cartridge. The phenomenon was precisely the same in the three cases."
"M. Maissin, who has succeeded M. Louppe at the Pont du Buis factory, states he was surprised to see the disorganized conditions of that factory. When there were signs of B powder decomposing it was sent back to the factory to be reworked—steeped in any alcohol until it contained 8%—and it was then to be put in special cases to be used at target practice. This, however, was customarily sent out as new, and not as reworked. As the powder which blew up the Liberte was sent from the Pont du Buis factory, it might now prove that although designated as 1906, it was many years older, and, as certain data on the duration of stability of reworked powder are not as yet available, the probable life of such powder could not be predicted.
This disaster has been made the opportunity by many papers, at home and abroad, of an attack on our—United States—powders. Even the disaster to the Maine is quoted, though it had not one pound of smokeless powder on board, the Mikasa also, though her explosion was admittedly caused by a fire and originated outside the magazines. The arguments as to the unwisdom of our using nitrocellulose powders have ail been answered satisfactorily by our Ordnance Bureau to the most persistent doubter, however prejudiced he may be.
The following few notes may show further where some of our differences between French and United States powders occur and why our powders are vastly more stable.
B powder is composed of two kinds of nitrocellulose, soluble and insoluble, of 11.60 and 13.23% nitration, respectively. Amyl alcohol was until very recently the solvent. The use of this has recently been the object of much adverse criticism, though the claim has always been made that its introduction doubled the life of the powder. The French have now given it up for the ethyl alcohol, and diphenylamine is used as a stabilizer.
The nitrocellulose is manufactured at one factory, the powder at another, a practice we have given up years ago as unreliable. The required stability test of nitrocellulose with potassium iodide starch paper at 65.5 ° C. is but 20 minutes. A very faulty purification could easily give this, and inspectors easily be hoodwinked.
The powder is made in ribbon form, and the variation in its ballistic qualities is made by variation in the percentages of the soluble and insoluble nitrocellulose in its composition, and not by varying the web thickness.
The processes used in French manufacture differ radically from those of the United States, particularly as regards attention to detail and thoroughness of purification in all the various steps. The stability of any nitrocellulose powder is absolutely dependent on the quality of the raw materials and on the thoroughness with which the details and purification processes are carried out step by step in its manufacture. The French process of drying is complicated, and includes water drying, which was abandoned, after experimentation, in the United States as injurious to the stability of the material. It seems strange that powder manufactured in the country of its inventor should not be as stable as a similar production of another nation, but such appears to be the fact.
There is in France at this time also a general complaint relative to the keeping qualities of sporting powders, the defects of which are attributed entirely to careless manufacture.
Thus we see that the stability of French nitrocellulose powders cannot be considered in the same light as ours. Over ten years ago stringent regulations were prepared by our powder board, it detailing exactly what purification must be given to our materials, and then to our powder in all its various processes of manufacture. These have been the cause of the high stability of our powders. It is seen that in France no one connected with the navy inspected the processes of manufacture, which inspection is here most thorough. In the majority of cases our powders are still stable after twelve years, and cases of spontaneous decomposition are practically unknown.
The damage to the Iena was caused by the explosion of a large bulk of black powder set off by the fire caused by the spontaneous ignition of a cartridge of B powder, while that to the Liberte is due to the explosion of shells overheated by fire started on board by ignition of similar powder.
In the French service shells are stowed in the same compartment with the charges, and not separately as with us.
The danger due to the spontaneous decomposition of our powder is this, and only this, that it will set fire to the woodwork, fixed ammunition boxes, supports of tanks, bins, etc., now existing in our magazines, and thus start a fire which might rapidly become uncontrollable, heating our shells in adjacent rooms to the bursting point. Remove the woodwork and combustible material from our magazines and a disaster like that to the Liberte will be impossible in our service.
?