Selected literature

Contents - Previous - Next

Grain fumigation is a broad subject. What may be considered basic information and principles have been presented in this chapter. There is a steady flow of publications on this subject in national and international journals and in the official government publications of many countries. Valuable and specific information is snmetimes only available in instruction booklets and brochures put out by fumigant manufacturers. The following literature references should be consulted as a guide to sources of more detailed information, on recirculation and forced distribution:

Berck (1975); Brook and Redlinger (1954); Brown and Heseltine (1949); Cotton and Walkden (1951); Holman (1960b); Howe and Klepser (1958); Lindgren and Vincent (1960); Monro (1956); Philips (1952, 1955, 1956, 1957b); Phillips et al (1953); Phillips and Latta (1949, 1953); Redlinger (1957b); Sergeev et al (1965); Storey (1967, 1971a, 1971b); Whitney and Kenage (1960).

11.Fumigation and controlled atmosphere storage

In addition to the poisonous gases that are used for pest control, the normal gases of the atmosphere can be altered to achieve control. The use and manipulation of natural components of the atmosphere, e.g. oxygen, nitrogen and carbon dioxide, to preserve food is referred to as "controlled" or "modified" atmosphere storage. Controlled atmosphere techniques are widely used in the storage of perishable commodities such as fruit, vegetables, cut flowers, etc. to retard ripening and reduce spoilage from micro-organisms. Also, they will control some insects in these products (Morgan and Gaunce, 1975; Aharoni et al, 1981). The most extensive use of controlled atmospheres for insect control is on grain and similar commodities. Here the atmospheres are modified by removing the life-supporting oxygen or by adding high levels of carbon dioxide.

Although the principle of modified atmosphere storage has been used since antiquity, e.g. in hermetic storages, a number of procedures have been developed in recent years to replace the normal atmosphere of a storage for the purpose of controlling pest organisms. In many respects the practice of using modified atmospheres for insect control is closely related to fumigation. Gas-tight enclosures are required, many of the procedures are closely related and the problems are often similar to those found in fumigation. When carbon dioxide is used, it is applied as a fumigant and it functions in a similar way.

Controlled atmosphere procedures are an appropriate substitute for the fumigation of some commodities because the gases involved do not leave harmful residues and often the atmospheres provide superior conditions to normal storage in air. In some cases the two procedures may be used in a complementary way to increase effectiveness of a treatment; carbon dioxide increases the toxicity of a number of fumigants to insects (Jones, 1938; Kashi and Bond, 1975).

Since sound pest management should, where possible, promote programmes that integrate appropriate control procedures and minimize the use of toxic chemicals and also since many of the requirements for fumigation are similar to those needed for controlled atmospheres, a brief account of controlled atmosphere procedures is given here.

Basic requirements

Controlled atmosphere systems depend on either depletion of oxygen to asphyxiate organisms or the addition of carbon dioxide to act directly and kill them. In these treatments the new atmospheres are maintained for an adequate period to kill all stages of the organism, and they should have no adverse effect on the commodity. To achieve this the treatment requires:

- a storage structure capable of containing the gas;

- a source of suitable gas or a means of producing the required atmosphere;

- a method of maintaining the atmosphere for the required period of time;

- a method of aerating to remove the altered atmosphere after the treatment.

It should be noted that the controlled atmospheres, which are toxic to pest organisms, are also dangerous to humans and precautions are necessary to ensure that no one is exposed to them without special protection. Although nitrogen itself is non-toxic to humans, the absence of oxygen or the presence of high levels of carbon dioxide is lethal.


Oxygen deficient atmospheres are produced by flushing a storage with nitrogen to displace the normal nitrogen-oxygen atmosphere. Liquid nitrogen from tanks may be used as a gas source (Banks and Annis, 1977). Exothermic inert atmosphere generators that consume the oxygen to leave principally nitrogen have been tested and show promise for insect control (Storey, 1973). These generators burn propane or other hydrocarbon fuel to give an atmosphere of less than 1 percent oxygen with about 10 percent carbon dioxide and 89 percent nitrogen. Oxygen can also be removed by the metabolic activity of micro-organisms and insects in hermetic storages, thus producing an atmosphere where insects cannot survive.

For complete insect control the level of oxygen must be maintained below 1.2 percent for one week, at temperatures above 35°C, or more than 24 weeks at 15°C (Table 15).

Grain temp. (°C) Exposure time (weeks)
15 24
18 15
20 6
23 4
26 3
30 2
35 1

Source: Banks and Annis, 1977.


Insects are generally killed more rapidly by carbon dioxide than they are by lack of oxygen. A concentration of 6n percent carbon dioxide will give over 95 percent control of most stored grain insects after a fourday exposure at 27°C or higher (Jay, 1971); however, longer periods are needed for complete kill. Banks (1979) suggested that an initial level exceeding 70 percent carbon dioxide and maintained above 35 percent for ten days is appropriate for complete insect control at temperatures above 20°C.

The carbon dioxide gas is applied to storages from a vessel of liquid carbon dioxide with appropriate vaporizers and pressure regulators to control flow rate (Jay and Pearman, 1973). Carbon dioxide in the form of dry ice has also been used for the treatment of grain in freight containers (Banks and Sharp, 1979; Sharp and Banks, 1980) and in conjunction with fumigation of grain with methyl bromide (Calderon and Carmi, 1973).


Structures used for controlled atmosphere treatments must have a high degree of gas tightness for the process to be effective and economical. They must be of sound construction and suitably modified so that gross gas loss through apertures, such as ventilators, open eaves and imperfections in the fabric, is prevented. Changes in temperature, atmospheric pressure and wind forces can have a pronounced effect on gas loss from the storage structure. For storages of 300 to 10 000 tonnes capacity a gas tightness that corresponds to a decay time of 5 minutes for an applied excess pressure drop of 2 500 to l 500, 1 500 to 750 or 500 to 250 Pa in a full storage has been found satisfactory (Banks et al, 1980). This specification corresponds to a whole area of no more than 1.0 cm²(0.16 in.2) in a 2 000 tonne storage.

Gas is added to maintain concentrations at the required level in a storage over the entire exposure period. One of the main reasons that gas loss occurs is the diurnal temperature variation in the neadspace of storage structures. This is less in concrete structures than in unprotected metal bins. Diurnal temperature fluctuations can be reduced by insulation, painting with a highly reflective white paint or by placing a false roof on the bin to leave a ventilated air space next to the permanent roof.


In structures other than welded steel, e.g. concrete, bolted or riveted metal, sealing of the entire fabric of the structure may be necessary. Bolted metal bins can be sealed by treatment of each lap joint and bolt location with silicon rubber sealant, thixotropic acrylics or by application of liquid envelope, "cocoon" type, polyvinyl chloride coatings. Concrete surfaces also may need to be coated with a good sealant that will prevent yes loss and protect the concrete from high levels of carbon dioxide. The permanent sealing of a 16 000 tonne capacity shed for fumigation or modified atmosphere storage of grain has been described by Banks et al (1979).

In bins that are very tightly sealed, some precautions are necessary to avoid unusual stresses on the structure caused by external or internal pressure changes. To prevent such changes in a bin that is sealed for a controlled atmosphere treatment, a pressure relief valve must be installed. If the bin will withstand the pressure, an operating level of + 1 000 Pa (4 in water gauge) appears to be suitable. Lower levels can be used but they should not be less than + 250 Pa. A simple U-tube valve of 8 cm (3 in) I.D. tubing with a liquid trap provides a convenient and fool-proof venting system (Banks and Annis, 1977).


Structures can be tested for gas tightness using a pressure test system (static testing or pressure decay testing) or a procedure using carbon monoxide as a tracer gas. The pressure decay system is satisfactory for routine testing of bins and is given in some detail here. Further information on testing of storage structures for gas tightness may be found in the publications by Banks and Annis (1977) and Banks (1982).

To test a sealed structure by the pressure decay method, air is introduced by the gas introduction system from a blower capable of producing 6 m/min (212 ft³/min) at 2 500 Pa (10 in water gauge). Pressure differentiels can be measured conveniently on a portable water gauge similar to that shown in Figure 43. If a pressure of 2 500 Pa cannot be reached with such a blower, there may be too much restriction between the blower and the bin or the bin is less gas tight than required (e.g. 3.1 m/min at 2 500 Pa for a 2 000 tonne bin). Care should be taken to ensure that pressure within the bin does not exceed the engineering limitations of the bin or of the sealing that is used. When a pressure of 2 500 Pa or the design limit is achieved, air input is cut off in such a way that no air is lost back through the blower. Pressure decay on the water gauge is then related to the period of time involved.

Procedures for establishing controlled atmospheres

Controlled atmosphere storage can be viewed in two phases - the "purge phase" where the normal atmosphere is replaced with the prescribed atmosphere and the "maintenance phase" where the atmosphere is maintained for the desired period of time (Banks and Annis, 1977).


Purge Phase

Liquid nitrogen supplied directly from a road tanker is passed through a heat exchanging facility where it is vaporized and brought to ambient temperature. It is then passed through a flow meter (e.g. "Rotameter") into the gas introduction system of the bin. Five cm (2 in) I.D. PVC drainage piping can be conveniently used to carry the gas. A flow of 3 m/min (106 ft³/min) has been found to be suitable for purging of bins from 300 to 7 000 tonne capacity, although this rate may be substantially increased (e.g. to 8 m3/min) in bins fitted with aeration ducts modified to introduce gas. However, when gas in introduced directly into the bin at the walls, an increased input rate may be less efficient at lowering the oxygen tension. In cases where a low efficiency distribution system is used, pockets of air may remain which are not purged directly but in which the oxygen is removed through the slower diffusive and convective forces. In these instances a slower purge rate allows these processes to occur and is not wasteful of gas.

FIGURE 43. Water gauge for static testing or for pressure decay(Friesen, 1976) measurements.

A vent of at least 50 cm²(10 in) must be left open in the roof during the purge to prevent dangerous pressure build-up.

It is important that the heat exchanging facility is adequate to bring the gas close to ambient temperature (within 2°C). Cooling of the grain near the introduction point may result in moisture migrating to the cold area on long storage and is also detrimental to the insecticidal efficiency of the process. At 3 m /min (106 ft³/min) three Forced draught heat exchangers in parallel using 0.4 kW (0.5 h.p.) fans, with an atmospheric exchanger downstream in series, have been found to be just as effective. If icing occurs downstream of the exchangers, the input flow must be reduced.

Gas input must continue uninterrupted until the headspace has been reduced to about 1 percent O2. Purging in the grain mass occurs generally by the passage of a sharp front through the grain and direct displacement of the interstitial air. In the headspace where free gas mixing occurs, decay is exponential. When the headspace has reached 1 percent O2, the purge may be terminated, the top vent and introduction ports closed and the maintenance input of gas commenced.

The quantity of nitrogen required for the purge is strongly dependent on the ratio of the volume occupied by the bulk to the total storage volume (the filling ratio). Barley has a higher porosity and will require more nitrogen than wheat. In one trial with barley where the filling ratio was 0.80, a volume of 1.9 m³ N2/tonne was consumed. At 3 m3/min the purging of a 2 000 tonne bin of wheat takes about 12 hours.

Maintenance Phase

After termination of purging, gas input at a lower rate is continued to maintain the atmosphere. The input rate required is determined by the bin capacity, its degree of gas tightness and the weather. At present it is best to adjust the maintenance rates by systematically reducing the flow of gas until the atmosphere is just maintained. At the correct rate, with grain temperatures exceeding ambient ones, the oxygen tension at the base of the bin will rise slightly during the day and fall to about 1 percent at night. The headspace tension should remain low. If the grain temperature is below ambient, or if the atmosphere contains more than 3 percent CO2, effects will be seen in the headspace, not at the base.

The exposure periods recommended for different grain temperatures with 1 percent O2 in nitrogen have been given above in Table 15. These figures are tentative and subject to revision when further laboratory studies are completed. Sitophilus oryzae is one of the most tolerant grain pests in low oxygen atmospheres, and if there is any doubt about what species are present, the exposure for this pest should be used. In commodities where the pests are Tribolium spp. and Oryzaephilus spp. only, shcrter exposure tames, half of those given, may he used. Currently available data for Rhyzopertha dominica are insufficient for an exact recommendation, but it is considerably more susceptible than Sitophilus oryzae and should thus be controlled by the exposures given. Exposure times should be based on the minimum temperature within the grain, not average temperatures. (For further details see Banks and Annis, 1977).


A number of inert atmospher3e generators with capacities ranging from 1.4 m³ (50 ft³)/h to 2 800 m (100 000 ft³)/h and capable of producing atmospheres with less than 1 percent oxygen are available on the market. Storey (1973) conducted tests in a 40 x 5.5 m (130 x 18 ft) silo containing 544 tonnes of wheat with two generators, each one capable of generating 420 m³ per h of inert atmosphere (<0.1 percent 02' 8.5 - 11.5 percent CO2 plus N2). Using 3 cm (1.5 in) drain feeder pipes connected at the bottom of the bin, he was able to reduce the oxygen level throughout the bin to 1 percent or less in 48 hours. The oxygen deficient atmosphere moved upward through the grain mass with very little intermixing with the normal atmosphere at a rate about 2.4 m (8 ft)/h. In another bin that was purged with the inert atmosphere prior to filling with grain, the oxygen level throughout the bin, including the headspace, was reduced to less than 1 percent within 24 hours.

Similar tests were carried out by Navarro et al (1979) in welded steel bins containing about 1 200 tonnes of wheat with a generator producing exhaust gases at a rate of 144 m /h. When the generator was in a recirculation arrangement so that the gases were blown in through the roof of the silo and drawn back into the converter through a pipe at the base, the average oxygen concentration was reduced to 0.2 percent in 60.3 h (purge phase). During the maintenance phase, which lasted 21 days, the oxygen concentration was maintained below 2 percent by intermittent operation of the generator for a total of 19.5 h.

It should be pointed out that the moisture content of the wheat was affected by purging with this atmosphere, particularly in the upper layer at the region of introduction of the gases. A plastic sheet was placed below the point of gas entry to prevent moistening of the grain.


Grain Storages

Three procedures for establishing high concentrations of carbon dioxide in large silo-type bins have been tested by Jay (1980). These are based on introducing the gas at the top of a filled storage, at the bottom of the storage or with the grain stream during filling. A procedure for introducing the gas at the base of the storage with subsequent recirculstion has been found to be effective in controlling a natural infestation of insects and is deemed to be commercially feasible (Wilson et al 1980).

For the latter treatment, liquid carbon dioxide supplied in cryogenic tankers is vaporized in a heat exchanger and the gas is diluted with air to give approximately 80 percent carbon dioxide and 20 percent air. The gas stream is maintained above 30°C with a superheater and introduced at the base of the bin through a 75 mm diameter iron pipe. When the carbon dioxide concentration at the top of the bin reaches a constant level, gas input is stopped and the atmosphere so established is recirculated through a 50 mm diameter plastic duct leading from the bin apex to the introduction port at the base. With the gas tightness standards specified above, this procedure will give a concentration of carbon dioxide >70 percent initially and this can be maintained at > 35 percent for ten days without requiring additional gas.

Freiqht Containers

Successful tests have also been carried out with carbon dioxide for disinfestation of freight containers loaded with wheat. In trials on gastight general purpose containers filled with commodity and dosed with 37 - 55 kg of crushed dry ice spread over the goods, plus 44 kg in blocks packed in insulated boxes, Banks and Annis (1980, 1981) found that the carbon dioxide level remained over 35 percent for a period of seven days or more and killed all of the test insects.

Tests have also been made for treatment of grain under gas-proof sheets with carbon dioxide (Banks and Annis, 1980).

Termination of treatment

After a controlled atmosphere treatment is terminated, the modified atmosphere is gradually displaced by entry of air from outside. The rate at which this happens depends on a number of factors, such as gas tightness of the storage, contents of the storage and the weather, and it may be increased by making larger openings or by active ventilation with a blower. A 2 000 tonne bin with 5 cm (2 in) diameter openings at the top and base may reach acceptable oxygen levels of 16 percent from the established 1 percent in less than two weeks (Banks and Annis, 1977).

Care should be taken to avoid structural damage from the reduced pressures caused by rapid emptying of a tightly sealed storage. Steel bins have been severely damaged by rapid removal of grain without sufficient venting.

NOTE: Before emptying tightly sealed bins, open access doors or vents to allow adequate air to enter as grain is removed.

Safety precautions

Adequate precautions should be taken when working in areas close to controlled atmospheres or on entering storages that have been treated to avoid any harmful effects. Nitrogen atmospheres containing less than 14 percent oxygen or more than 5 percent carbon dioxide may be dangerous to human life. Personnel entering a nitrogen atmosphere containing less than 10 percent oxygen may collapse without warning and become uncon scious. Carbon dioxide produces respiratory discomfort, lightheadedness and nausea, and unconsciousness may occur in less than five minutes in 9 percent CO2. In any case, where unconsciousness or respiratory distress occurs through exposure to a controlled atmosphere, the victim must be taken immediately to fresh air.

Portable oxygen monitors or carbon dioxide analysers should be available on the work site as well as air-line or self-contained breathing apparatus. Gas masks with canisters provide no protection against low oxygen or high carbon dioxide atmospheres. All enclosed working areas close to controlled atmospheres should be well ventilated and gas reservoirs should be kept outside if possible (Banks and Annis, 1977).

Problems associated with modified atmosphere storage

In sealed systems, moisture may migrate and condense to produce problems, particularly in grain above 12 percent moisture content. High carbon dioxide or low oxygen atmospheres can inhibit mould and toxin formation and preserve germination under moist storage conditions. However, spoilage will occur once the modified atmosphere is removed unless the grain is dried or processed within a short time (Banks, 1981).

Once the modified atmosphere is removed the commodity is open to infestation unless otherwise protected.

Grain kept under a modified atmosphere may develop taint; at higher moisture levels (> 16 percent) this taint may be difficult to remove (Banks, 1981; Shejbal, 1979).

High levels of carbon dioxide may cause structural problems in reinforced concrete storages by reducing alkalinity of the steel (Hamada, 1968). The significance of this reaction in grain storage is currently under investigation.

Choice of treatment

When a controlled atmosphere procedure is going to be used, the choice between using the nitrogen (oxygen deficiency) method or the carbon dioxide method may be determined by several factors. The amount of time available for the treatment, the suitability of the structure for holding the gas, the availability of the gases or equipment for producing them and the overall cost of the operation are all important. Carbon dioxide generally kills insects faster than nitrogen (oxygen deficiency) and requires a less stringent standard of gas tightness (Jay, 1980). The comparative cost of controlled atmosphere treatments will depend on the availability and cost of the gases, the amount required and transportation costs, as well as the costs of equipment and labour.

Fumigation, controlled atmospheres and forced aeration

The choice of treatment to be used for controlling insect infestations will vary with the type of problem and the relative merits of the treatment. Fumigants can usually be used effectively in storages prepared for controlled atmosphere treatments, as the standards of gas tightness that are required are generally more stringent for the controlled atmospheres. Also, the quantity of material required, the application costs and the exposure time are usually much less for fumigants than for controlled atmosphere treatments. On the other hand, controlled atmosphere procedures avoid the use of chemical pesticides, leave no harmful residues and can provide superior storage of grain. A combination of fumigation with controlled atmosphere procedures may have some potential, as the effectiveness of fumigants is enhanced by carbon dioxide.

In addition to controlled atmospheres, the forced aeration of grain is closely allied with insect control procedures and may be an integral part of a pest management programme. Calderon (1972) suggested that "a sensible use of ambient or chilled air for aeration of grain offers new possibilities in many parts of the world for preservation of grain without (or with very little) use of chemicals." The use of fumigants and controlled atmospheres together with forced drying procedures and aeration should be considered as complementary conservation methods that form part of an overall pest management programme.

It must be pointed out that controlled atmosphere procedures are in early stages of development and progressive changes are likely to be made with further experience and research. A select bibliography on controlled atmosphere and aeration procedures up to 1981 is given below and may be referred to when extensive use of these procedures is planned.

Selected references


Bailey (1955); Banks and Annis (1977, 1981); Banks et al (1979); Banks and Sharp (1979); Calderon and Carmi (1973); Hyde (1962, 1969); Jay (1980); Jay and Pearman (1973); Kruger (1960); Navarro et al (1979); Press and Harein (1966); Press et al (1967); Rannfelt (1980);'Sharp and Banks (1980); Shejbal (1980); Storey (1973); Vayssiere (1948).


Armitage and Burrell (1978); Burges and Burrell (1964); Burrell (1967); Calderon (1972); Friesen (1976); Hearle and Hall (1963); Holman (1960a); Navarro (1976); Oxley and Wickenden (1963); Williams (1973).

Contents - Previous - Next