Gard recently published an article focusing on the toxicity of phosphine gas and the potential danger this posed to the crew when agricultural bulk cargoes and forestry products were undergoing in-transit fumigation with aluminium phosphide.
In that article Gard briefly mentioned another potential hazard that could arise at the initial stage of the process – phosphine gas explosions.
Dr Nicholas Crouch has written an article focusing on the risk of explosion in port and the steps that could be taken to reduce the risk.
Aluminium phosphide, the precursor compound from which phosphine gas is generated, is available in different forms and can be supplied in aluminium bottles as tablets of about 3g each or as pellets of about 1g. Alternatively it can be supplied in fabric ‘socks’ or strips of cojoined paper sachets. These enable the fumigant residues to be removed easily in a situation where it was undesirable to have fumigation residue remaining in the cargo. Generally, and as a rule of thumb, Degesch state that one 3g tablet of aluminium phosphide formulation will generate about 1g of phosphine gas.
Usually the quantity of aluminium phosphide applied per hold is calculated based on the volume of the hold rather than the quantity of cargo it contains, with a common target concentration of phosphine deemed to be effective of 1g of gas per cubic metre, although sometimes higher concentrations were used.
When arranging an ‘in-transit’ fumigation, the lead fumigator calculates the required quantity of aluminium phosphide on a hold-by-hold basis and the fumigation crew then applies this across the surface of the cargo stow, sometimes using pipes to place a proportion of the fumigant beneath the surface of the stow.
Once exposed to water vapour in the air the aluminium phosphide reacts, generating phosphine gas and aluminium oxide as a by-product. “It should therefore be appreciated that all the phosphine for the entire hold is generated on the surface or near the surface of the stow where a considerable proportion will accumulate in the headspace before dispersing/diffusing into the cargo”, wrote Dr Crouch.
The overall fumigation operation can be divided into four distinct phases with these being:
- Phosphine gas generation and accumulation in the headspace of the hold as the aluminium phosphide reacts with moisture,
- Diffusion of the gas into the interstitial spaces between the kernels of grain or soybeans leading to reduced concentration of phosphine in the headspace,
- The contact period during which any live insects are exposed to the fumigant, and
- Release and dispersal of any remaining fumigant by ventilation following the designated period of contact and prior to discharge.
Phosphine and a related compound, diphosphine, had long been recognized by chemists to be highly reactive and quite unstable chemicals with both being prone to self-ignition. This was especially so for diphosphine, which could be formed when there was an imbalance in the amount of aluminium and phosphorus in the aluminium phosphide tablet or pellet formulation.
The manufacturers of aluminium phosphide formulations include ammonium carbamate in the mixture as this decomposed on exposure to water vapour to generate carbon dioxide and ammonia. These gases were believed to suppress the likelihood of ignition by decreasing the inflammability of the phosphine. However, at concentrations of 1.8% or above in air, phosphine is inflammable and will support ‘combustion’ producing dense white fumes of phosphorous pentoxide which is a severe respiratory tract irritant due to formation of orthophosphoric acid on contact with water, such as in the lungs. In the presence of an ignition source, such elevated concentrations of phosphine will ignite. In an enclosed space, such as the headspace of a hold, could cause an explosion.
Potential sources of ignition were:
- Hot work on the hatch covers,
- Exposure to temperatures above phosphine’s self-ignition temperature,
- Direct inadvertent wetting of the fumigant tablet; this can rapidly produce very high phosphine concentrations and, because the phosphine producing reaction is exothermic, temperature conditions above the self-ignition point of phosphine.
- Unwanted production of diphosphine from the fumigant formulation as discussed above.
- Sparks from, say, static electricity in hatch cover structures or direct inadvertent metal on metal contact,
In the many cases that Brookes Bell investigated it was usually impossible to identify the source of ignition with absolute certainty, but there were generally common features which point to them being in some way part causative.
- Explosions generally occur within the first 8 to 24 hours, presumably due to the high phosphine concentrations in the headspaces during this period, but Brookes Bell was aware of one explosion which occurred after 29 hours,
- Use of high doses of aluminium phosphide/phosphine which leads to higher initial phosphine concentrations during the first phase of the fumigation.
- High prevailing ambient temperatures and high humidity, both of which increase the rate of phosphine generation,
- Intense tropical sunshine which can heat the hatch covers possibly exceeding the autoignition temperature of the phosphine,
- Surface application of the fumigant may result in higher headspace concentrations of phosphine, and
- Comparatively small headspaces above the cargo stows which lead to higher phosphine gas concentrations during the initial phase of the fumigation.
Fumigations on board ships were entirely the responsibility of the fumigators, over which Masters and crew have no control or influence. Consequently, once an “unsafe” fumigation has been set there was nothing the crew could do to minimize the risk of a fumigant explosion from occurring.
Nevertheless, wrote Dr Crouch, “as a general safety precaution in the early period of fumigations in ships’ holds, we would recommend that as little time as possible is spent on deck and close to the hatch covers by the crew during the first 24 to 36 hours after the fumigation has been set”.
After this period, it is reasonable to expect that most of the aluminium tablets will have reacted to release the phosphine gas with a proportion of that gas diffusing into the cargo, thereby reducing the concentration of phosphine in the headspace of the hold and therefore reducing the risk of explosion.