At 1435 hrs on 4th June 2004 the 37272dwt chemical tanker NCC Mekka was in the process of tank cleaning
when a low pressure explosion occurred in tank 1CS which had previously carried parrafinic solvent.
This was followed by another explosion in the adjacent tank 1CP which was fully loaded with ethanol.
The deck was fractured in several places and the escaping ethanol caught fire,
the fire spreading all the way aft to the deck house.
The vessel was en-route from Santos to Aratu (Brazil).
The crew extinguished the fire by 1505hrs, using the vessel’s foam monitors,
and managed to bring the vessel to the Rio de Janeiro roadstead.
One able seaman and the Boatswain were badly burned and subsequently died.
Water ballast tank no. 1 starboard was fully charged,
while water ballast tank no.1 port was empty.
Due to the explosion, cracks were formed between these tanks,
and water ballast therefore ran over to the port tank.
This contributed to a list of the ship to port of about 5 degrees.
Cracks also formed between tanks 1CS and 1CP and between those two tanks and the water ballast tanks.
This caused ethanol from 1CP to run down into the water ballast tanks and over into 1CS.
There were bulges, but no punctures or cracks in the outer bottom and hull.
Following emergency consultations with the company, the master decided
to perform an emergency discharge (ethanol and water ballast) into the sea.
Water-driven rotary tank-washing machines were being used to wash tanks
which had contained parrafinic solvent which from a product data sheet
subsequently provided by the shipper was found to have a flash point of -40 deg C.
Prior to the explosion the tanks had been subjected to a half-hour cold seawater wash,
followed by a two-hour hot seawater wash at about 60 deg C.
About 200 litres of detergent had then been added to the tank via the closed sampling inlet of a Butterworth hatch.
2 to 3 c.metres of fresh water was added via the cargo line system
and a three hour washing programme commenced, using the fresh water/detergent at ambient temperature.
This fresh water/detergent wash was in progress at the time of the explosion
Investigators considered various sources of ignition,
including mechanical friction from the cleaning machines and/or the deepwell pump,
electrical short circuit in tank monitoring systems and structural damages.
All these were excluded based on information available and examinations carried out after the explosion.
The conclusion was that there must have been a charged atmosphere in tank 1CS,
and that a static discharge had taken place inside the tank.
What led to the discharge was not clear
but it was postulated that a static charge could have been created from the detergent-loaded water wash.
The presence of human error could not be excluded,
but nothing in particular was specified that could indicate such error.
Although the vessel was fitted with an oil burning inert gas generator
the cargo tanks were not inerted at the time of the casualty.
At the maritime inquiry, the chief mate stated that the inert gas system on board (which was based on oil combustion)
was unacceptable to the charterers because of too low purity.
However no explanation was given as to why nitrogen was not used for inerting or why, in
the absence of an inerting medium the cargo was not rejected.
A data sheet (commodity information) that the NMI found available on board
at the maritime inquiry did not specify whether paraffinic solvent
(which is included in the category “naphtha” in MARPOL Annex I) was flammable,
nor did it specify the flash point of the substance.
There was no mention of whether such information had been requested. \
(The data sheet gave the product name as paraffinic solvent, shipping name: naphtha solvent).
The NMI received another data sheet at the company office on 25 June.
In this sheet it was stated that the flash point was -40° C,
and that the substance was flammable.