Section 9 - Chemical control methods fumigation technology

Contents - Previous - Next

Fumigation technology in developing countres - Report of discussions and recommendations
Storage pests and their control manual
Fumigation for the control of insect pests in storage
Code of practice for storage and fumigation of bagged grain
Gas monitoring and detection systems


Fumigation technology in developing countres - Report of discussions and recommendations


Phosphine is applied to commodities as a solid formulation containing aluminium or magpoisons. This means that they must be applied as gases or as liquids which readily vaporise to produce toxice gas. This is in contrast to contact or stomach insecticides which poison insects through direct contact or ingestion in the liquid or solid state. Fumigants play a major worldwide role in the protection of stored grain from insect pests, but are particularly important in tropical and subtropical regions where conditions are most favourable for insect development. They have the particular advantage over other insecticides in that they can penetrate to all parts of a bulk or bag stack and, if used properly, can produce a complete kill of all stages of insect pests.

Fumigants which are applied in the gaseous state, such as hydrogen cyanide or methyl bromide, have used for many years in the developed countries. Certain fumigants applied as volatile liquids, e.g. ethylene dichloride, ethylene dibromide, carbon tetrachloride and carbon disulphide, have also been commonly used as grain fumigants, often as mixtures. It was only about 30 years ago that fumigation, mostly with methyl bromide or ethylene dichloride/carbon tetrachloride, began to be carried out in developing countries. The introduction of phosphine, generated by exposing solid formulations containing aluminium or magnesium phosphide to air, greatly simplified application procedures so enabling fumigation to be used more readily in developing countries. Unfortunately the expansion in the use of phosphine was not always accompantied by appropriate training in the techniques of application needed to ensure a total kill of the insect infestation.

Recently there has been increasing evidence showing the development of resistance by certain major pests to phosphine. Levels of resistance in some strains of certain species have increased to such an extent that it is difficult to achieve succesful results using a recommended dosage rate and fumigation procedure. There is much evidence to suggest that such resistance is associated with the use of poor fumigation techniques over many years. These difficulties with phosphine have been accompanied by a gradual withdrawal of the older alternatives which are either much less convenient to use or are now suspected of having carcinogenic properties. The scope for the introduction of completely new fumigants which might be used as phosphine replacements is necessarily very limited due to the small number of chemicals possessing the necessary properties, and the high cost of producing all pre-registration data now required by pesticide registration authorities.

The objective of the seminar was to discuss the present situation and to identify ways in which the organisations represented by GASGA and other bodies throughout the world might respond to it. The role of phosphine as a fumigant is crucial to developing countries and if its continued effectiveness were to become doubtful, without the existence of suitable alternatives a grave situation would be created.


Phosphine is applied to commodities as a solid formulation containing aluminium or magnesium phosphide which reacts with atmospheric moisture to liberate phosphine. Formulations based on magnesium phosphide generate phosphine more rapidly than those containing aluminium phosphide. Phosphine is also liberated more rapidly from both types of formulation under ambient conditions of high relative humidity and temperature. The manufactured products contain ingredients which retard this reaction or produce carbon dioxide which prevents spontaneous combustion of phosphine.

Four main types of formulation are currently available for the generation of phosphine; (1) flat or round tablets generating 1 g of phosphine, (2) flat of round pellets generating 0.2 g of phosphine, (3) sachets generating 11 g of phosphine and (4) 'plates' generating 33 g of phosphine. Of these only the fiat tablets and pellets are manufactured in developing countries. Each of these has particular characteristics in relation to the rate of generation of gas. Pest control organisations in developing countries need to consider which type is most suitable for local conditions, including also ease of application and safety, packaging and storability and cost when deciding which formulation to use (Halliday, page 19). If tablets are produced locally there may be no choice and cost will generally be the most important factor.

Techniques have been developed by the United States Department of Agriculture (USDA) for the fumigation of bulk grain in tankers and bulk dry cargo vessels using phosphine. These techniques are claimed to be both safe and effective under the prevailing circumstances. The margins for safety and efficacy are very narrow, however, and the techniques should only be used under favourable conditions. Such fumigations may be carried ourt while the vessel is at sea thus reducing the need to delay shipments. In-transit fumigation of grain has proved particularly valuable to US grain exproters and is now also being carried out on shipments of bulk grain from France, the Federal Republic of Germany and Argentina. The procedure developed by the USDA has been endorsed by the International Maritime Organisation (IMO). The possibility of some survival of insects during such fumigations was recognised but it has to be accepted that the current inspection and handling procedures often permit no alternative to in-transit ship-board. fumigation when cargoes were found to be infested. Ship-board fumigations should be confined to bulk grain in tankers and bulk dry cargo vessels. No techniques have yet been developed for the safe and efficient fumigation of cargo types of ships, break bulk cargoes and 'tween deckers and fumigation in these vessels should not be attempted.

There are three methods by which flat bulks of grain are treated with phosphine

(1) Probing phosphine formulations throughout the bulk with or without covering with gas-proof sheets.

(2) Placing the fumigations on the surface of the grain: fumigation of bulk grain in this way is only satisfactory if carried out at temperatures exceeding 15°C and it is essential that adequate dosages be applied. There is great concern at the low dosages that are now being recomended for use in developing countries. Dosing at levels substantially above the theoretical minimum is needed for full control, especially where there is difficulty in obtaining the high standards of sealing necessary to enable the exposure periods to phosphine required for control.

(3) 'Spot' fumigation, in which only limited portions of the grain bulk are fumigated by probing formulations into areas which are identified as being infested was deplored. Spot fumigation with phosphine has only a 50% chance of success at the first attempt and could lead to a gradual build up of resistance.

Experiments in Togo demonstrated the successful phosphine fumigation of small rural storage structures containing sorghum grain, despite considerable leakage of gas, with application rates of 6 g phosphine/m³. It was considered that phosphine could be used for the disinfestation of grain developing countries at village or small trader level in appropriate circumstances. The need to ensure safety was emphasised, however, especially as storage installations are often intermingled with living accommodation. This can only be achieved with good sealing which is necessary in any case if efficacy is to be achieved and the danger of build-up of resistance to phosphine is to be avoided. Also there is a need for small packs of phosphine formulations so as to avoid the danger of traders breaking large packs and retailing tablets in unsealed containers.


Sensitive, accurate, reliable, rapid and protable methods of analysis are needed for monitoring concentrations of fumigations in the field. This is necessary both for assurance that adequate concentration - time products are obtained and to avoid exposure of personnel to hazardous concentations.

The most common method used in the field for the rapid determination of phosphine and methyl bromide concentrations is that of detector tubes. These contain cotton wool impregnated with a chemical reagent which changes colour as a gas sample is drawn through the tube. A measure of the concentration of fumigant is indicated by the length of the decolourised cotton wool. Detector tubes are calibrated directly for this purpose and can be used to give approximate indications of fumigant concentrations within a period of 5 min or so. Detector tubes are not sufficiently accurate for the determination of phosphine concentrations in experimental programmes.

Other methods for monitoring fumigant concentrations depend on their physical rather than their chemical properties. A meter measuring methyl bromide concentrations by thermal conductivity has been in use for 30 years. Portable infra-red gas analysers for the determination of both phosphine and methyl bromide have been used for 10 years or more, while instruments based on sonic analysis or interferometry have also been developed. None of these, however, has found any widespread application except for research purposes.

The most accurate method of analysis for both methyl bromide and phosphine remains that of gas liquid chromatography (glc). It is difficult to use this technique in the field, however, except in mobile laboratories. Such facilities are generally only available to sophisticated research organisations.

Two methods for phosphine analysis which have recently been developed were demonstrated. They combine the accuracy of glc with the other requisites of field test equipment. They have significant potential, not only for research purposes but also as a more accurate replacement for detector tubes. The first of these, developed by the Tropical development and Research Institute (TDRI), is based on the reaction of phosphine with mercuric chloride in a modified cell connected to a conductivity meter which provides a direct digital reading of phosphine in parts per million (Harris, page 56). The method is simple to operate and has been successfully used in a wide range of field conditions in Africa and Asia. Accuracy, determined by comparison with glc, is about ± 5%, including sampling errors. The equipment may be assembled from standard components, it is cheap and can be carried in a small camera case.

The other method developed at Laboratoire Denrées Stockées in France uses the reducing activity of phosphine in an electrochemical cell. When gas containing phosphine is admitted to the cell an oxydoreduction reaction is induced and produces electric current in proportion to the concentration of phosphine present. This can be converted into a direct reading of phosphine in parts per million on a digital meter. The method has been tested in field trials in France and has provided a reliable method of measuring phosphine concentrations over a range of conditions. It may be possible to adapt it to give a continuous recording of phosphine concentrations.


Many aspects of phosphine toxicity to insects are still incompletely understood but it is known that oxygen is needed for the gas to be effective and different species, and developmental stages of the same species of insects, vary enormously in their susceptibility. Some stages are so tolerant that they are able to continue normal development in the presence of quite high concentrations of gas. In so doing they pass from the tolerant stage to one of greater susceptibility. If the fumigation area is adequately sealed death of these stages occurs shortly after a susceptible stage is reached. Obviously, with an actively increasing population comprising adults and immature stages, there is much to be gained if the exposure pelod can be increased to span the tolerant stages in the life cycle. The length of the exposure required is temperature dependent because it is governed by the rate of insect development. Increasing concentration of gas over shorter periods is unlikely to improve control in these circumstances.

In tests on adults or diapausing larvae where there is little or no progressive development, the period of exposure to phosphine is still more important concentration. For larvae in diapause there is much variation in the minimum time of exposure required for death among individuals in tests at high concentrations. Minimum periods to achieve 50 or 90% kills are relatively long compared with other stages. For this reason, efficacy of phosphine, unlike other fumigants is not determined by the product of concentration and period of exposure. Formulae equating concentration of phosphine and exposure periods cannot, therefore, be expressed in terms of the product product producing a constant mortality (as for other fumigants such as methyl bromide). They must also take account of minimum gas concentration and exposure period needed to achieve required levels of mortality.

Superimposed upon the basic toxicity pattern is a narcotic threshold, encountered at higher concentrations; this is a state assumed by insects in which they become dormant and are not killed by phosphine. The concentration range over which this effect is observed may increase at higher temperatures. At concentrations inducing narcosis some insects survive for longer periods than at lower concentrations, but the protection conferred is temporary. Unexpected survival after relatively long exposures is also encountered when insects experience a gradual rise in gas concentration from zero (Reichmuth, page 88). This occurs in practice as gas is released from a proprietary formulation. This effect may also be linked with narcosis because, when concentrations rise to 1 or 2 mg/l in about a day, narcotic levels are encountered within a few hours of dosing. At present phosphine toxicity data has largely been obtained using constant concentrations and more work is required on changing levels.

The mode action of phosphine is still uncertain, however, some interesting results have been obtained from uptake studies comparing resistant and susceptible strains. Examples of phosphine resistance so far studied have not been associated with narcosis but physical activity and respiration have appeared unaffected for a substantial initial period of the exposure. Clear evidence has been obtained in Rhyzopertha dominica for a mechanism which actively excludes the gas and which greatly delays the accumulation of poison in the tissues. Four other species have a similar mechanism. Investigations are underway to determine if exposure conditions can be manipulated to overcome this mechanism.

Resistance to phosphine has been detected in insects from a number of developing countries. Tests on immature stages have shown that some resistant strains of several species now require doses as high as the naturally tolerant Sitophilus spp. for control. Resistance has been detected in strains of six species of beetle collected in the UK. It was observed that no discriminating dose test has yet been developed for species of moth infesting stored products.

Work in Australia has shown that resistance can be induced in most strains of insect species by exposure to phosphine in the laboratory. The relative importance of time of exposure and concentration level varied in different strains, although time of exposure was always the more important factor. The minimum level of concentration required for the fumigation to be effective was raised in resistant strains, indicating a greater ability for detoxication. The narcotic threshold was also higher in resistant strains.

No instance has so far been detected in which resistance to phosphine has developed to the extent that infestation cannot be controlled by recommended procedures. However, it is vital that fumigations with phosphine only be carried out where exposure to lethal concentrations can be guaranteed for adequate periods. This implies either good conditions of sealing or the use of novel procedures in which multiple application of fumigant can be carried out. There is also a need to update information on the incidence of resistance in countries where. phosphine is used.


The FAO global survey or insecticide resistance carried out in 1972-73 by Champ and Dyte (1976)* showed evidence of phosphine resistance by a number of important species of insect pests of stored products, and pointed out the potential for further development of resistance to phosphine.

The first positive confirmation of fumigation failure due to phosphine resistance occurred in Bangladesh in 1981. This could be related to poor fumigation practice in which leaky godowns had been fumigated with inadequate doses of phosphine over many years. Substantial levels or resistance were noted in R. dominica, Cryptolestes ferrugineus and Tribolium castanceum. Highly resistant strains of C. ferrugineus have been found on commodities imported into the UK from the Indian subcontinent, suggesting that the resistance noted in Bangladesh was not an isolated occurrence.

TDRI is now carrying out a survey of the current geographical incidence of resistance to phosphine. Resistance has been observed in strains of insects from Pakistan, Nepal, Bhutan, Sri Lanka, Botswana, Mali and Tunisia. Strains from Ethiopia, Liberia, Nigeria, Zimbabwe, USA, Burma, St. Lucia and Thailand were found to be non-resistant. Evidence was presented to suggest that some of the resistant strains had originated in other countries and had been transferred on imports. This and similar work to update current knowledge of resistance to phosphine is most valuable and should be continued and strongly encouraged.

The causes of resistance development have been examined in detail. The area where resistance appears to be most prominent is in the Indian subcontinent where the fumigation of whole stores without proper sealing is commonly practiced. This is the same cause as the observed resistance in Bangladesh, and fumigation procedures similar to those used in Bangladesh are widely practiced in India, Pakistan and Nepal. Such practices result in repeated underdosing and inadequate exposure periods which cause a gradual increase in the proportion of the total population showing resistance.

It was recognised that fumigation of whole stores is likely to continue because the simplicity and cost of the technique made it attractive. Fumigation of such stores can be improved without any great increase in cost or technical complexity by better sealing, the use of a phosphine formulation which releases gas more slowly and by multiple application of formulation. The need for further research into the development of suitable fumigation techniques, taking more account of meteorological factors such as temperature variation and wind was recognised. The importance of developing such techniques for whole-store fumigation is paramount.


The use of carbon dioxide as a fumigant for grain stored in large bulks has been demonstrated in the USA, Australia and other countries. Such treatment is effective if concentrations of about 90% are maintained. For this reason it is likely to be economic only where carbon dioxide is available at very low prices. Interest in this technique is being shown by countries in Southeast Asia where cheap carbon dioxide is available.

The use of carbon dioxide mixed with small quantities of phosphine or methyl bromide was described. Such techniques have been shown to be effective and were likely to be of most use for the fumigation of bulk grain in silos and containers.

The use of oxygen-free ('inert') atmospheres to control infestation of bulk grain was described. The atmospheres may be produced by combustion and feeding the exhaust gas into the structure. They have the advantage over carbon dioxide in that they can be readily produced on site and are thus independent of industrial production of gases. The effective use of inert atmospheres requires a high degree of airtightness as the presence of even very low concentrations of oxygen will allow insects to survive. Such air tightness is most easily obtained in flexible plastic enclosures. Whilst there is potential for such systems there may be difficulties in applying them in most developing countries.

Liquid fumigants might be reconsidered to a limited degree as a replacement for phosphine. Potentially they are of particular use for the fumigation of small quantities of bulk grain, spot fumigation in large bulks and the localised fumigation of milling equipment. Most traditional liquid grain fumigants, e.g. ethylene dibromide or ethylene dichloride administered with carbon tetrachloride, are suspected of possible carcinogenicity and other toxic effects. A new mixture (80% trichloroethane, 20% methyl bromide) was discussed and it was hoped it would meet the new, more stringent, requirements for mammalian toxicity and still be as effective as the former fumigants.

Hydrogen cyanide was at one time extensively used for the fumigation of grain and other commodities but has been almost completely superseded by methyl bromide or phosphine. There might be some scope for the reintroduction of hydrogen cyanide in particular circumstances where phosphine or methyl bromide cannot be used.


1. Specifications for Phosphine Formulations

Notional regulatory authorities which operate pesticide registration schemes should require suppliers of phosphine formulations to provide information, with supporting experimental data, on the total yield and rate of release of phosphine under specified conditions of temperature and relative humidity which approximate to those that might be expected during use.

The authorities should also require that the standard of packaging of phosphine formulations should be such as to ensure that unopened containers can be safely stored without deterioration for at least 2 years. The size of packages should be appropriate to the local pattern of use.

On the basis of most currently recommended dosage reates and methods of application for phosphine-generating products, and in the light of current knowledge, exposure periods of less than 5 days should be discouraged in recommendations on labels or in product information sheets supplied by manufacturers or their agents. Labels on phosphinegenerating products should not contain any information that suggests that phosphine is suitable for use in fumigation enclosures that are of a lower standard of gas tightness than that required for other fumigants.

2. Fumigation on Ships

Grain exporters and shippers should note that techniques and safeguards have been developed for the in-transit fumigation of bulk grain in tankers and bulk dry cargo vessels. However, it is recognised that margins for safety and efficacy are too narrow for this technique to be widely recommended and it should be confined to those situations where good results can be assured. In particular, in-transit fumigation in ships' holds cannot yet be recommended as a suitable approach to disinfesting grain or similar commodities for developing countries. No such techniques have been developed for break bulk cargoes and 'tween deck ships and it is strongly recommended that intransit fumigation of such cargoes or in such vessels should not be attempted at all pending further research and the development of a safe and effective methodology. Guidelines have been published by the IMO on this subject.

3. Fumigation of Large Bulks of Grain

Organisations storing large quantities of grain in bulk should undertake fumigations only under good conditions of sealing. Phosphine fumiga ions of bulk stores should be attempted only on the total bulk, and not in localised areas of it, so as to ensure full control of insect infestation and to avoid the development of resistance.

4. Resistance of Insects to Phosphine

The Group noted the scientific studies on the toxicity of phosphine, its mode of action and the mechanisms of resistance in progress at several centres in developed countries. These studies are vital for a thorough understanding of problems of resistance which have occurred in practical fumigation. It is most strongly recommended that these studies be continued and increased and, additionally, extended to cover the higher temperatures prevailing in the tropics.

The Group recognised the value of the FAO worldwide survey of insecticide resistance carried out in 1972-73 by Champ and Dyte which clearly indicated the potential for a phosphine resistance problem. This has now been shown to exist in many geographic areas. It is considered of the greatest importance and most urgent that the current status of the degree and geographical extent of resistance to phosphine be assessed. It is further recommended that donor agencies:

(1) consider this programme a project of high priority;

(2) consider how they might contribute to the identification of countries, regions and situations where insecticide resistance, particularly fumigant resistance, is serious and increasing;

(3) implement programmes to improve fumigation technology where these are most urgently required.

5. Application Technology

Government departments, parastatal organisations, commercial pest control operators and all those with responsibility for stored grains should note that the primary cause of control failures leading to the development of resistance is poor fumigation techniques. These are exemplified by inadequate dosage, insufficient exposure time, poor and improper sealing and failure to recognise and understand the significance of prevailing ambient conditions.

The Group recognised that the most effective method of ensuring satisfactory standards of control of infestation in bagged grains and flat bulks in nonfumigable warehouses is by fumigating under undamaged gas-proof sheets. Developing countries need to be advised on specifications for fumigation sheets, and their proper maintenance and repair and storage when not in use.

The Group also recognised that fumigation of whole stores will continue to be carried out but that many of the observed practices in the sealing of the building and application of the fumigant are inadequate. Specifications need to be provided for the adequate sealing of stores to be constructed, having regard to the cost and availability of local materials and skills.

Studies are also needed on the feasibility and cost of sealing existing buildings. Where it is not possible to realise an effective total seal means should be developed to measure the effectiveness of the attainable seal and application techniques be developed appropriate to the condition of the store.

The need for adequate training and a proper status for pest control operational personnel, at all levels, should be recognised by senior management if the work is to be given its proper priority and good results are to be obtained.

6. Detection and Monitoring of Fumigant Concentrations

New and relatively inexpensive methods of measuring phosphine are now available. Pest control operators, marketing boards other responsible authorities should be actively encouraged to use these methods to monitor fumigations for which they are responsible. Detector tubes are now available to monitor concentrations of many industrial gases but the limitations of accuracy of those used to measure fumigant concentrations should be recognised.

It is a matter of concern that recently accepted threshold levels of methyl bromide cannot be detected using an halide lamp and new wethods for detection in the field at these concentrations are urgently needed.

7. Alternative Fumigants

The Group recognised the danger of there being only two widely used fumigants, phosphine and methyl bromide. It also recognised a need for the development of alternatives and took note of carbon dioxide, used exclusively or as a mixture with other gases, methyl chloroform/ methyl bromide mixtures and nitrogen. It was considered, however, that the complexities of logistical support and application needed would, at present, inhibit acceptance of these alternatives in developing countries. Nevertheless additional research in this area would be most beneficial in the long term. It should be noted that there may still be a role for hydrogen cyanide in some situations.

8. Fumigation on Farms

Fumigation in farms and small village stores should not be encouraged unless safe and effective treatments can be ensured. This will entail the provision of fumigant in packs of appropriate size and good sealing of structures and commodities being fumigated. Such conditions can rarely be achieved under present circumstances.

Contents - Previous - Next